Communications system and method

ES 2 265 043 T3

DESCRIPTION

Communications system and method.

The invention relates to aspects of a communication protocol and in particular to aspects of a communication protocol for a multi-user wireless communication system.

The invention further relates to a communication system using such protocols, a method for wireless communication between an access point and a plurality of subscriber units, and a radio signal for transmission between an access point and a station unit. subscriber.

Description of the prior art

In a multi-user wireless communication system, and in particular in a fixed wireless access (FWA), a simple access point (AP) communicates with a number of subscriber units (SU). Since APs are typically more expensive than SUs and more expensive to locate, an AP should preferably be able to communicate with as many SUs as possible. In a wireless system, however, the bandwidth is usually limited and therefore it is very important to use the bandwidth as efficiently as possible.

Subscriber traffic usually includes data and voice traffic. Voice traffic is relatively easy for a system operator to handle, because a voice channel can be fragmented almost arbitrarily on the radio link to accommodate the communications protocol used by the system operator. Data traffic is more difficult to handle because it involves the provision of widely varying amounts of bandwidth to different subscribers at different times according to their individual requirements.

In conventional systems, there are problems in efficiently packing the data traffic into the communications protocol of the system operator. For example, in a radio network it is advantageous to use a communication protocol having fixed length frames because it is easy for radio receivers to synchronize with such a structure.

An example is International Patent Application No. WO96 / 38930 disclosing an apparatus and method for establishing and maintaining communication paths in a wireless telecommunication system. The receiver of a subscriber terminal compares the code and phase of the master code sequence in the downstream signal (from the network) with the code and phase of the slave code sequence of the receiver. The receiver adjusts the phase of the slave code sequence until a match is obtained with the master code sequence. The downstream signal includes a monitoring channel having a power control signal, a code sync signal, and a frame alignment signal. The receiver at the exchange terminal monitors the signal transmitted upstream (to the network) by the transmitter at the subscriber terminal and provides changes in the code sync signal so that the transmitter is in sync with the receiver. The receiver monitors the downstream signal to identify the framing signal and establishes the downstream communication path when two successive framing signals are identified. In the acquisition mode during the establishment of the downlink communication link the downlink signal is transmitted with a high power level and a low transmission speed. In the sleep mode after the downlink communication link establishment, the downlink signal is transmitted with low power level and low transmission speed. In traffic mode, upon request for a wireless communication transmission, the downstream signal is transmitted with a high power level and a high transmission speed.

There are problems, however, in the efficient use of bandwidth within fixed-length frames when users send different types of traffic.

The present invention attempts to overcome these and other limitations of conventional systems.

Summary of the invention

The invention provides various aspects of a communications protocol as defined in the accompanying independent claims, to which we will now refer. Preferred or advantageous features of the invention are defined in the dependent sub-claims.

In a first aspect, the invention therefore provides a communications protocol for a radio system in which, on the downstream signal, frames of variable length are dispersed around a regular, periodic pattern of synchronization pulses. Each sync pulse contains a pointer to the beginning of the next frame. The Su's can therefore be synchronized to the regular sync pulses and used to find the frame structure, while the system can efficiently pack the data into frames of variable length, according to user demand.

Each frame conveniently starts with a header. This not only describes the content of the frame, which has been programmed by the AP, but also a pointer to the beginning of the next downstream frame. The SUs that

ES 2 265 043 T3 continuously decode the downstream frames can therefore maintain synchronization without the need to refer to each synchronization pulse.

In a further aspect of the invention, the upstream signal carries variable length frames, in which the AP can efficiently schedule data transmissions from SUs. The content of the upstream frames is conveniently described in an uplink description within each header of the downstream signal. If required, for example if a frame is much longer than subsequent upstream frames, then one or more extra upstream signal descriptions can be programmed into the downstream frame. This ensures a conveniently efficient use of the upstream signal.

When an SU is initialized, or powered up, it can receive the downstream signal but needs training to acquire upstream channel details before it can effectively transmit on the upstream channel. In a further aspect of the invention, the AP regularly broadcasts basic information which, in combination with the SU observations of the downstream signal, conveniently allows the SU to transmit an initial training pulse. The AP gives the SU an opportunity to transmit this pulse within a wide training slot, which is wide enough to allow propagation delays between the AP and the SU. Each training pulse contains an address of the SU that sends it, and as such following successful reception of a training pulse within the wide training slot, the AP can send information to the SU to improve its subsequent transmissions. The AP may preferably program additional narrow training slots (which consume less upstream bandwidth than wide training slots) to allow transmission of training pulses by SUs that already have timing information to compensate for propagation delays. These training opportunities can allow SUs to follow changes in the upstream transmission channel over time.

In a further aspect of the invention, the AP can program contention slots, in which an SU that wants to transmit data can transmit a contention pulse. The AP may not know in advance which SU is transmitting a particular containment pulse, and therefore SUs pre-distort their containment pulses (preferably using information obtained during the training procedure) for reception by the AP. Conveniently, the communication system of the invention may be capable of receiving containment pulses without using an equalizer). The contention pulse identifies the SU sending it, and thus when a contention pulse is received successfully, the AP can enter into a dialogue with the SU regardless of the bandwidth requirements of the SU, and can finally program width upband for the SU as required.

Conveniently, the communication system of the invention uses cells in the asynchronous transfer mode (ATM) to carry both data and control information on upstream and downstream communications. This allows for very efficient packing into variable length frames.

In a radio communication system, channel conditions can lead to failure to receive transmitted information. Under these circumstances, an efficient confirmation procedure is desirable, to ensure that any lost data is retransmitted. A further aspect of the invention addresses this problem. When a SU receives the ATM cells programmed from the AP, these cells are numbered in sequence as in a conventional ATM transmission. If all cells are received successfully, the SU confirms its reception, but if the reception of any of the cells fails, the SU sends an acknowledgment message containing a sequence number to identify the first failed cell. The AP then only needs to reschedule the transmission of the following cells.In a further preferred aspect of the invention, the confirmation messages may contain a bitmap identifying the cells received successfully and the cells failed within the sequence of cells, so that the AP only needs to reprogram the transmission of the failed cells. Specific embodiments and the best mode of the invention

Specific embodiments of the invention will now be described by way of example with reference to the drawings in which;

Figure 1 illustrates variable length downstream frames scattered around physical layer sync pulses;

Figure 2 illustrates the use of offset pointers within sync pulses to locate downstream frames;

Figure 3 illustrates the header of a descending frame;

Figure 4 shows the use of an offset pointer between downstream frames;

Figure 5 illustrates the use of an offset pointer in a descending frame header to locate the structure of the ascending frame;

Figure 6 shows an extra uplink frame descriptor;

ES 2 265 043 T3

Figure 7 shows the use of a scroll pointer within the extra upstream frame descriptor of Figure 6;

Figure 8 illustrates the structure of a descending frame;

Figure 9 illustrates the use of the downstream event descriptor field within the downstream frame header;

Figure 10 shows the structure of a downlink frame including an extra uplink frame descriptor;

Figure 11 illustrates the structure of a downlink frame and the use of an extra uplink frame descriptor;

Figure 12 illustrates the upstream event descriptor fields contained within the upstream frame descriptors;

Figure 13 illustrates the riser structure;

Figure 14 illustrates the use of training slots on the upstream signal;

Figure 15 illustrates an upward training pulse transmitted by an SU;

Figure 16 illustrates the initial states of the SU registration protocol;

Figure 17 illustrates the SU training protocol;

Figure 18 shows the SU containment protocol;

Figure 19 is a block diagram of the structure of the AP and SU illustrating the use of VCI addressing;

Figure 20 illustrates the training protocol for SUs using narrow training slots;

Figure 21 shows the synchronization hierarchy of the communication system;

Figure 22 illustrates cell transmission and acknowledgments on the downstream signal;

Figure 23 illustrates the contention protocol for a SU that wants to transmit on the uplink;

Figure 24 illustrates the transmission and confirmation of cells on the uplink;

Figure 25 illustrates the hierarchy of recovery modes in the upstream cell transmission protocol; Figure 26 illustrates the down-pulse structures used in the communication system and Figure 27 illustrates the up-pulse structures used in the communication system.

System caracteristics

The embodiments of the invention described below are implemented in a fixed wireless access (FWA) system, in which as many as 1,000 subscribers, or users, can be served from a single access point (AP). In a sectorized cell system, each sector of an AP can serve as many as 1,000 subscribers. The system uses bi-directional frequency division (FDD) spread spectrum transmissions, such as code division multiple access (CDMA) transmissions. The downstream signals from the AP to the SU units carry both control information and data in virtual channels (VC) in asynchronous transfer mode (ATM) within a frame structure, as well as the uplink from the SU to the AP .

Physical Layer Synchronization

As shown in Figure 1, the downstream signal transmitted by the AP carries physical layer synchronization pulses (PHY-SYNC) 2 at intervals of exactly 10 msec. Figure 2 shows the structure of each sync pulse, comprising a correlation sequence of 112 PHY-1 symbols, a correlation sequence of 16 PHY-2 symbols, a 12-bit frame offset, and a parity check of 4 bits. The term symbols refer to the symbols of a spread spectrum communication system.

The PHY-1 and PHY-2 correlation sequences are used by each SU to hook into sync pulses and as sync and training sequences to set the receiver and modem parameters to match the transmission channel. To acquire synchronization, the control software in the SU receiver first sets the automatic gain control (AGC) level based on the SU modem output signal level. The SU modem and Media Access Control (MAC) then initiate automatic media control.

ES 2 265 043 T3 frequency (AFC), which has three different tracking modes. First, for example when an SU is turned on for the first time or if it loses the channel, the acquisition mode is used. In each sync pulse the correlation sequences are predetermined pseudo-random sequences, which are known to the SU. When the SU receives the downstream signal, the correlator output therefore produces maxima when the sequences are received. The acquisition mode uses the time offset between maxima of the correlator to initialize the AFC and thus acquire the descending channel.

Second, a coarse AFC mode positions the maximum correlation within one of the 64 time slots (T / 4) and adjusts the AFC until the correlation is maximum in the correct slot at all times.

Third, the fine mode adjusts the shoulders of the maximum correlation to be uniform.

Descending Frame Structure Location

The media access control (MAC) in the AP transmits control information and subscriber data within the frames, which are sandwiched between the physical layer sync pulses. A sequence of frames 3 and their arrangement around the synchronization pulses are shown in Figure 1. The frame length is variable and as such the sync pulses can appear anywhere between or within the frames.

Each SU can be locked to the physical layer using the sync pulses as described above. Each of these pulses 2 contains a 12-bit frame offset 4 as shown in Figure 2, which is a pointer to the beginning of the next downstream frame. The offset is given as the number of spread spectrum symbols (a symbol count) and allows each SU to locate the beginning of the next downward frame. Figure 2 illustrates the offsets 4 between three consecutive sync pulses and the start of each subsequent frame.

The header at the beginning of each descending frame is modulated using quaternary phase shift coding (QPSK). Figure 3 illustrates the structure of a descending frame header 6. It comprises a descending frame descriptor 8 followed by an ascending frame descriptor 10. The descending frame descriptor starts with the same correlation sequence of 16 PHY- symbols. 2 as carried by each sync pulse. This allows a correlator in the SU to positively identify the beginning of the frame. The synchronization sequence is followed by the 16-bit frame offset 12, which is the symbol count until the beginning of the next downward frame; that is, once an SU has been locked into the frame structure, in the meantime it can continue to decode the frame structure it can synchronize with the start of each descending frame without additional direct reference to the sync pulses. A descending description 14 follows offset 12 and describes the content of the frame. The downstream frame descriptor ends with a cyclic redundancy code (CRC).

Figure 4 illustrates the use of the descending frame offset (DL) 12 to refer from the header of a DL frame (frame #n) to the beginning of the next DL frame (frame # n + 1). Figure 4 shows only the DL frames from the MAC, omitting the sync pulses.

Rising Frame Structure Location

The upstream frames are not all the same length. Also, the upstream frames do not need to be the same length as the downstream frames, nor do they need to start at the same time.

As shown in Figure 3, the uplink frame descriptor 6 carried in each downlink frame header contains a 16 bit uplink frame offset 22 followed by an uplink description 24 and a 16 bit CRC. Figure 5 illustrates the offset of the upstream frame 22 providing a symbol count from the current downstream frame to the beginning of the next upstream frame 26. As in Figure 4, Figure 5 omits the sync pulses for clarity, showing only the MAC downlink and uplink frames.

The offset between the header of the downstream frame and the start of the next upstream frame will vary for each SU depending on the distance between the AP and the SU, due to propagation delays. In a preferred embodiment, therefore the offset of the upstream frame 22 is the offset experienced by an SU in the maximum range of the AP and each SU must add a delay to the offset depending on its own range from the SU.

If the downstream frames are significantly longer than the upstream frames at any time, for example due to the traffic load on the downstream and upstream channels, then providing only one upstream frame descriptor in each downstream frame can result in inconvenient low utilization. of the ascending channel. Under these circumstances, extra upstream frame descriptors may be inserted into the downstream frames to announce the start of the new upstream frames. Figure 6 shows the structure of an extra upstream frame descriptor, which is identical to the upstream frame descriptor 10 shown in Figure 3 that is part of a downstream frame header except that, like all upstream pulses, it starts with a PHY-2 correlation sequence to improve synchronization at the AP. Figure 5 showed the usage

ES 2 265 043 T3 of a simple upstream frame offset in a downstream frame header to indicate the beginning of the next upstream frame 26. Figure 7 is a similar drawing showing a much longer downstream frame (downstream frame #n ) carrying a first upstream frame offset 22 in its header and an extra upstream frame offset 30 between the data carried by the downstream frame. The extra upstream frame descriptor is always modulated using QPSK so that it can be demodulated by all SUs, even if it is enveloped by higher order modulations as described above.

Descending Frame Structure

Following the header, each downstream frame carries information pulses for SUs, which may include ATM cells or confirmation signals (see below). As mentioned above, the downstream headers are modulated using QPSK. Other information pulses can be modulated differently, for example using 16-QAM or 64-QAM, but in each frame these modulations must be in ascending order. Thus, the downstream frame #n in Figure 8 contains, following the QPSK header, confirmation QPSK signals (ACK) and ATM cells 16, amplitude and quadrature modulated ACK signals 16 (QAM) and 18 cells and then ACK in 64-QAM and 20 cells. Different SUs may have different modem capabilities. All must be capable of demodulating QPSK modulation but SUs can incorporate modems that have more advanced performance to demodulate 16-QAM or 64-QAM modulations. Consequently, it cannot be guaranteed that all SUs can demodulate the entire frame but they can all demodulate the QPSK header, including the downstream frame descriptor 8.

If a modem that is only capable of demodulating QPSK attempts to demodulate a frame that carries higher order modulations, as illustrated in Figure 8, the phase capture and decision feedback equalizer (DFE) will be out of phase synchronization. because an attempt to demodulate higher order modulations will generate a very high Symbol Error Rate, which prevents the DFE from operating. However, SUs that can demodulate only the lowest modulation used by an AP can maintain symbol synchronization using QPSK sync pulses.

As described above with reference to Figure 3, the downstream frame descriptor 8 carried in the header 6 of each downstream frame carries a PHY-2 training sequence, the downstream frame offset 12, a downstream description 14 and a 16-bit CRC.

PHY-2 is used to train the digital phase capture loop (D-PLL) of the return end of each SU modem and to load the feedback stages of the DFE. This may not always be necessary but will be required if the previous frame contained higher order modulation sections which caused some SUs to lose DFE / D-PLL synchronization.

The downward shift is the symbol count to the beginning of the next downward descriptor, which is the beginning of the next downward frame, as described above.

Downstream description 14 contains 21 8-bit characters and provides seven 24-bit descriptor fields for downstream events scheduled within the downstream frame. Figure 9 illustrates the four downstream event types that can be programmed numbered as type 0 through type 3. The type number is carried in two bits 32 in the header of each descriptor field. Type 0 is used to fill the descending description if the frame contains fewer than 7 scheduled events, and contains only zeros. Type 1 describes an acknowledgment event (ACK), for which the descriptor carries a two-bit modulation type QAM identifier and a twelve-bit ATM virtual channel identifier (VCI). Type 2 downstream event is the transmission of an ATM cell. In this case, the descriptor identifies the type of modulation (QAM), two bits, the ATM VCI, twelve bits, and the cell length, four bits. Type 3 describes an extra upstream frame descriptor; all of these have the same structure and therefore no information is required in the descending event descriptor field, which is padded with zeros. The order of the seven descriptor fields in the descending description matches the order of the descending events in the frame.

In total, there are three types of downstream events, cells (type 2), commits (type 1), and extra upstream frame descriptors (type 3). The upstream and downstream frame descriptors must always be in the lowest modulation order supported by the AP, eg QPSK. Clearly, therefore the extra ascending frame descriptor appearing in the last portion of a frame may violate the role that modulations must be in ascending order. SUs must still be able to identify the extra upstream frame descriptor by identifying the PHY-2 correlation sequence at its beginning.

Figure 10 illustrates a descending frame containing a header 6 followed by acknowledgments and cells in increasing order of modulation, and a descending frame further containing an extra ascending descriptor 28 between higher order modulations.

Figure 11 illustrates an upstream frame offset within a descending frame header in more detail, expanding the contents of a descending frame (descending frame #n) and the descending frame header as in Figure 8, and then shows the link provided by the upstream frame offset 22 to the beginning of the next upstream frame (upstream frame # n + 1).

ES 2 265 043 T3

Each downstream frame can carry pulses for SUs using different modulations. Therefore any given SU ​​can only be guaranteed to demodulate the QPSK DL frame descriptor. Each upstream frame descriptor contains a pointer to the beginning of the next UL frame so that the SU can locate it even if it cannot demodulate the higher order modulations.

As described above, the downstream frame header includes an upstream frame descriptor. This is implicit and does not need to be described in the descending event descriptor field.

Ascendant Frame Structure

There are no standard upstream frame structures, but in general each upstream frame will contain one contention slot, one upstream request, upward acknowledgments, and upward cells. Wide and narrow training slots and inquiry marks may also be present. The functions of these frame elements, or events, are described later.

As described above, the upstream frame descriptors are carried on the downstream signal, either within the downstream frame headers or as extra upstream descriptors. Figure 12 illustrates the structure of an upstream frame descriptor, and in particular upstream description 36. The upstream description comprises 21 8-bit characters, providing a variable number of descriptor fields for scheduled upstream events. There are 6 types of ascending events.

There are 7 types of descriptors, including type 0000 for filling the ascending description if required. Type 0000 contains the four-bit type number followed by four additional zeros. A query event is indicated by a descriptor type 0010; the descriptor carries the four bits of the type number 0010 and the twelve-bit ATM VCI. A contention event descriptor (slot) comprises a four-bit type number 0011, and indicates a contention slot in which SUs can compete for bandwidth access; this descriptor field contains the type number 0011 followed by four zeros. Type 0100 indicates a wide training slot, in which SUs can transmit broadcast training sequences; this descriptor field contains only the four-bit type number 0100. Type 0101 indicates a narrow training slot for SUs to transmit training sequences; this descriptor field contains the four-bit number type 0101 and a twelve-bit ATM VCI. Type 11 is a commit event; the descriptor field contains the two-bit type number 11, a two-bit modulation order identifier, and a 12-bit ATM VCI. Type 10 indicates the transmission of a cell by a SU; this descriptor field contains the two-bit type number 10, a two-bit modulation level indicator, a 12-bit ATM VCI, and a four-bit indication of the cell length, followed by four zeros.

These fields within the ascending description specify the contents of the next ascending frame, the beginning of the frame indicated by the ascending frame offset preceding the ascending description. Figure 13 shows a portion of an upstream frame. The upstream cells are sent in pulses from each SU that you want to transmit. The pulses have been programmed using cell descriptors in the ascending description (see Figure 12). Figure 13 shows pulses 38, 40 from SU # 1 and # 2. The pulses are separated by a guard time to prevent collisions. The cells are conventional ATM cells except that they have no need to carry the VCI, which has already been transmitted in the cell descriptor. An implicit upward request 42 starts each pulse, and is followed by cell number 44. Each upstream request starts with a PHY-2 training sequence, to allow the AP to resolve the phase uncertainty and train its DFE feedback cells, followed by a 4-bit field that indicates the number of frames to transmit (NUM), a Twelve-bit VCI and 16-bit CRC. Indeed, the VCI and NUM fields are superfluous and could be omitted because, as described below, the AP has already assigned an ATM VCI and a number of cells to the transmitting SU and can identify the VCI of the following cells and the identification of the SU from the position of the cells within the upstream frame. However, the transmission of the entire uplink request improves the synchronization of the AP and thus suitably increases the reliability over the uplink.

Ascending Training Mechanism

When an SU is first powered on, or first registered with the AP, it may not know the distance between itself and the AP. The SU can receive and lock onto the downstream signals but the transmission delay originating from its range from the AP affects the offset it sees between the downstream and upstream signals. The initial timing offset to be accounted for for its transmission delay may be referred to as the offset pre-delay.

Pre-compensation slots, or wide training slots as we refer to them in Figure 12, are programmed relatively infrequently on upstream frames. This is because they need to have a length twice the maximum propagation delay between the AP and any SU plus the length of the training pulse to be transmitted by the SU, and therefore consume significant bandwidth. Figure 14 illustrates a rising frame containing a wide training slot 46 and also shows the wide training slot in more detail. Figure 15 shows the structure of an upward training impulse 48.

ES 2 265 043 T3

To allow pre-delay compensation, the AP MACs program a wide training slot, using the up-frame descriptor, and each SU that requires delay compensation transmits an upward training pulse into the wide training slot. When this is done, the SU assumes that the propagation delay is zero. Consequently, the AP receives each training command from the SU at the time after the start of the wide training slot equal to twice the propagation delay between the AP and the SU (since the delay affects both up and down links). . As shown in Figure 15, each upstream training pulse includes the identity of the sending SU (SUID) 50, and as such the AP can inform each SU of its propagation delay.

When an SU transmits an upward training pulse, it may not have previously received any feedback from the AP to allow it to calibrate its transmit power or equalize to compensate for the channel. To try to ensure that the AP can decode the training pulse upward, the SU therefore uses an estimated transmit power and pre-distorts the training pulse as follows.

The SU can receive the downstream signal, and can therefore generate a received signal strength indicator (RSSI). To evaluate your initial transmit power, you also need to know the transmit power of the AP. It is broadcast by the AP on a regular basis, as described below. To pre-distort the training impulse upward, the SU evaluates the descending signal and assumes that the characteristics of the ascending and descending channel are the same. Pre-distortion techniques such as those described in US Patent No. 6031866 or International Patent Application No. PCT / GB00 / 00589 can be used.

As shown in Figure 15, each upward training pulse begins with the one hundred and twelve symbol PHY-1 and sixteen symbol PHY-2 correlation sequences, followed by the sixteen bit SUID. The correlation sequences are known to the AP and therefore can be decoded relatively easily, even if the SU power and pre-distortion evaluations are imprecise.

After initial delay compensation training, the SU should not need to use wide training slots. Regular upward retraining can be programmed by delay compensated SUs within the normal upward structure using narrow training slots 47. A delay compensated SU can send an upward training pulse into a narrow training slot, for use by the AP for feedback information to the SU to improve its equalization and transmit power control.

Ascending Training Protocol

Figure 16 illustrates the AP broadcast transmission carrying information required by the SU to attempt its registration, if desired, followed by programming a wide training slot using the upstream frame descriptor. The broadcast information includes the identity of the AP or service provider, the maximum delay (cell range), the transmit power of the AP, the CNR fading margin, and the SUID space. An SU can be located so that it can receive signals from more than one AP or in more than one AP sector, but it can only register for communications with one of these. When hooking onto a downstream signal, receiving the identity of the AP or service provider thus prevents its misregistration attempt.

Figure 17 illustrates the synchronization protocol using a wide training slot. When such a slot has been programmed as shown in Figure 16, if a SU wishes to train it sends a training pulse upward into its wide slot. The AP tries to software decode the impulse. If it fails, or if there was a collision between upward training pulses from two or more SUs in the training slot, then the SUs are reported using the AAL5 broadcast VCI. The broadcast message communicates to all of those SUs in a numbered frame specific to the failed training. Upon receiving such a broadcast signal, the SU knows that training failed and therefore sends an additional training pulse in a subsequent wide training slot, after a random period of rest.

If the AP successfully decodes an upward training pulse, it sends a broadcast message containing the SUID of the SU along with the power control correction and delay compensation information. This gives the SU the correct transmit power, timing, and valid SUID. The AP then sends 64 correlation results on the new defect management VCI (SU-VCI = 0). This gives the SU the upstream channel.

During the training procedure, the AP then assigns a SUID to the SU. The initial SUID included by the SU in its training sequence is a random number. After the upward training pulse has been successfully decoded, the AP assigns a SUID to the SU for use in further communications. Matching Strategy

A SU only has to communicate with one AP while an AP, in implementation, may have to manage up to 1,024 SUs. It is desirable to achieve efficient medium access control, with short transmitted pulses and contention on the uplink. A real-time equalizer strategy would inconveniently require too many training symbols at the beginning of each upward push. The implementation therefore uses off-line equalization with regular retraining and has the majority of the complexity spread across SUs. Of

ES 2 265 043 T3 In this mode, the initial training phase using wide training slots allows the SU and AP to resolve the ascending and descending channels. Upon receiving upward flow training pulses, the AP preloads the correct post-cursor stages for each SU. Each SU below pre-distorts its next scheduled broadcasts to reduce pre-cursor distortion and decrease noise. For upstream contention (which is not programmed for individual SUs), as described below, each SU totally pre-distorts its containment impulse so that the AP does not require an equalizer, or any inherent knowledge of which SU is sending a particular containment drive.

The matching strategy is illustrated in more detail in Figure 18, which shows that the AP only has to fully match training pulses, which are relatively infrequent. As shown in Figure 18, the AP broadcasts the downstream signal and the SU trains the downstream channel offline. Using a DFE and estimating an initial transmit power as described above, the SU pre-distorts a first training pulse that is transmitted in a wide training slot. The AP decodes the off-line training pulse and transmits upstream channel details back to the SU, using the SUID of the training pulse. This allows the SU to learn the characteristics of the upstream channel and improve its pre-distortion for further transmissions. When an SU transmits the following contention pulses, as described below, the AP does not know that the SU is sending the contention pulse before it arrives. The SU therefore transmits contention pulses using total pre-distortion so that the AP does not need to use an equalizer. When an SU subsequently transmits a programmed up pulse, as described below, the AP knows which SU is sending the pulse and can therefore decode it using a DFE preloaded with the feedback stages of the individual SU. The SU therefore transmits programmed up pulses using linear pre-distortion (pre-cursor removal).

SUID / VCI ratio

Each AP and each SU contains an ATM cell switch / multiplexer that passes ATM cells from one receiver port to another within the MAC, within a control module (RAMP), or outside the wired side transmission port. These options are numbered 3, 2, 1 respectively on the AP and SU block diagrams in Figure 19.

A default VCI is used for each SU for RAMP control. This VCI is recognized only on wired ports and at power-up has the same default values ​​for each unit, which can be reassigned once the SU is under the local control of either an Access Point Concentrator (APC) or the User Premises Equipment (CPE). A 16-bit VCI (v) mask is used to select the VCI space for the AP, or for an AP sector in a multi-sector system. Over the air VCIs are a subset of 12-bit VCIs. The following example illustrates the use of the VCI mask to generate the VCI address for different portions of the system, as illustrated in Figure 19. Example

<td>VCI</td><td> 1011</td><td> 0011</td><td> 1001</td><td> 0110</td><td>VCI over AP wired port</td>

<td>mask</td><td> 1111</td><td> 0000</td><td> 0000</td><td> 0000</td><td>system mask (configurable)</td>

<td>mask (v)</td><td> 1011</td><td> 0000</td><td> 0000</td><td> 0000</td><td>value of the sector mask, i.e. VCI for this AP and its SUs</td>

<td>VCI &amp; f</td><td> 1011</td><td> 0000</td><td> 0000</td><td> 0000</td><td>VCI &amp; f</td>

<td>VCI &amp; v</td><td> 1011</td><td> 0000</td><td> 0000</td><td> 0000</td><td>VCI &amp; f = VCI &amp; v, therefore this VCI is for this sector</td>

<td>f Denied</td><td> 0000</td><td> 1111</td><td> 1111</td><td> 1111</td><td></td>

<td>VCI &amp; f Neg.</td><td> 000</td><td> 0011</td><td> 1001</td><td> 0110</td><td>produces “VCI on air” -a + b, where</td>

<td>to</td><td></td><td> 0011</td><td></td><td></td><td>SUID (for this sector it is 4 bits wide)</td>

<td>b</td><td></td><td></td><td> 1001</td><td> 0110</td><td>SU-VCI</td>

<td>v + b</td><td> 1011</td><td> 0000</td><td> 1001</td><td> 0110</td><td>VCI over wired SU port</td>

(NB SUID standalone)

Broadcasting

When a SU locks onto transmissions from an AP for the first time, it cannot tell if that AP is on its own network and has no knowledge of physical layer parameters such as the AP's transmit power or cell radio. required for initial up-power control and delay compensation. A broadcast VCI is used to allow unregistered SUs to learn about the network and the physical layer. At intervals, the downlink contains AAL5 cells over a predetermined broadcast VCI. Next, as illustrated in Figure 16, the AP sends broadcast information about the VCI 0000 0000 0000 (VCIAAL5). The information comprises the maximum delay from the AP to the SU (the range of the AP cell), the SUID space used in the cell, the MAC address of the AP and / or the address of the service provider operating through the AP, the transmit power of the AP and the carrier-to-noise ratio (CNR) fading margin in dB.

ES 2 265 043 T3

Upward Periodic Retraining

Since the system of the embodiment employs off-line equalization, changes in the channel may not be tracked by the SUs. Regular retraining is desirable. This can be initiated either by the AP programming a particular SU for retraining or by the SU autonomously retraining in a wide upward retraining slot. Figure 20 illustrates the procedure for scheduled retraining.

If the AP decides that the uplink that a SU is degrading or if it has not received uplink communications from a SU for more than a predetermined period, it schedules an uplink training event by sending an uplink frame descriptor of type 0101 (training slot narrow) identifying the SU. The SU next sends a training impulse upward into the narrow ascending training slot. The AP waits to decode the impulse by software, eliminating the channel. In this case, the AP broadcasts the corrected power control information and the correlation samples return to the SU using the AAL5 management VCI. The SU has now corrected its transmit power and upstream channel parameters. If the AP cannot decode the upward training pulse, then it informs the SU accordingly, and the SU will re-initialize and attempt to recover communications.

Synchronization Recovery

Each SU aims to maintain synchronization by tracking the correlation sequences and frame deviations on the downlink. Each downstream pulse that contains a scroll pointer also contains a CRC for error control. If a CRC check fails, then the SU will be unable to use the scroll pointer to find the next frame. In such a case, the SU needs to go back to the previous level of the frame hierarchy to resynchronize, as illustrated in Figure 21. Thus, if a CRC, or a parity check fails when decoding a downsync pulse , the SU searches again for the next correlation peak generated by the synchronization sequence. If a CRC check fails when decoding a downstream frame descriptor 8, the SU again looks for the offset in the downstream sync pulse. If the CRC check on an upstream frame descriptor fails, the SU searches again for the next downstream frame descriptor.

Down Traffic Scheduling

When ATM cells are sent to a SU, for example to transport data directed to the SU or control information from the AP, the AP schedules the cells into the downstream signal as illustrated in Figure 22. In downstream description 14 to the frame in which cells are sent, the AP inserts a cell descriptor 52 of type 2 (see Figure 9). The cell descriptor contains the modulation level to use for the cells, which will depend on the modulation capabilities of the SU, a 12-bit VCI address and the length of the message (number of cells) to be sent. The VCI contains the SUID for the SU to which the cells are sent, and as such after receiving the cell descriptor, the SU expects to receive the cells later in the frame. The AP then sends the cells 54 as programmed in the downstream frame. The cells are conventional ATM cells except that they do not need to carry the VCI, which has already been transmitted in the cell descriptor. The SU receives the cells and sends an acknowledgment signal 56 on the uplink.

As shown in Figure 22, upward confirmation starts with a mapping sequence of sixteen PHY-2 symbols. It then carries an eight-bit sequence number (SEQ) 58 and a sixteen-bit bitmap (MAP) 60, which indicates to the AP if any of the cells sent on the downlink have been lost or not received properly. . ATM cells sent over the downlink were numbered sequentially using the SEQ field. The SEQ 58 field of the upward confirmation signal carries the SEQ Number of the first cell in the sequence that was not received properly. Thus, if 8 cells were sent and the fourth, fifth, and seventh were not received properly, the SEQ field in the up confirmation signal would carry a SEQ with the value of 4. Bitmap 60 then carries a bitmap of the next cells in the sequence, a value of 1 indicates a successfully received cell and a value of 0 indicates an improperly received cell. Thus, in the example given above, the bitmap field would carry bitmap 0101, indicating that the fifth and seventh cells have not been received. (In an alternative embodiment, the bitmap may end with the last cell that was received improperly, implying that subsequent cells were received successfully. The bitmap in the example above would then be read 010. In a second alternative, the Last cell received improperly would be similarly omitted from the bitmap, giving the shorter bitmap 01 in the example). Upon receipt of the upward acknowledgment, the AP schedules forwarding of any cells identified in the upward acknowledgment as not received. If no confirmation is received, the AP schedules forwarding of all cells. If a confirmation indicates that all cells have been received, the AP takes no action.

It is important that the system can recover from errors in the downstream process. At each stage of the downstream access state machine (sending downstream programming, sending cells, receiving confirmation) there is a possibility that a pulse will be lost or corrupted. The confirmation procedure allows you to recover how the AP will reprogram and resend the cells until it has received a successful confirmation.

ES 2 265 043 T3

Ascending Access Request

An SU can gain access to the upstream transmission by two methods; containment and inquiry. Snooping is used only for real-time services. Both methods result in an inquiry from the AP to the SU for its bandwidth location, as contention does not address the details of the bandwidth, but only a SUID.

Figure 23 illustrates the containment procedure. If a SU wishes to send cells over the upstream transmission, it first sends a fully pre-distorted upstream contention pulse with (approximately) the correct timing and transmission power in a narrow contention / training slot, as described above . The containment pulse contains a PHY-2 correlation sequence followed by the SUID 64 of the SU and 6 parity check bits. The AP waits to decode the contention pulse by software. If it fails, or if there was a collision between pulses sent by two SUs in the same training slot, then the AP takes no further action. If the SU receives no response from the AP, it assumes its contention has failed and retries contention after a random idle period.

If the AP successfully decodes the upstream contention pulse, it confirms the SU request by means of a POLL 66 in the upstream frame descriptor 10 of the next frame. This is a 0010 type event descriptor as shown in Figure 12. POLL 66 carries a 12-bit VCI, which the SU uses to transmit its upstream access request. This upstream request 68 contains the number of cells 70 required by the SU and the VCI provided by the AP.

Ascending Cell Programming and Confirmation

Once the AP knows a SU's request for uplink access, it schedules a slot on the uplink for the SU to send the cells. Figure 24 illustrates this procedure. The AP uses a cell descriptor field (type 10 in Figure 12) in an upstream frame description to schedule an upstream slot for the SU to send its cells. The SU then transmits the cells with linear pre-distortion, preceded by an additional upstream request 72. The upstream request is identical to the upstream request sent by the SU during contention and snooping but, in this case, it is used only by the AP for synchronization reasons.

Following the reception of the cells, the AP sends a downward acknowledge signal 74 (following a type 1 descriptor field in the downstream description 14, see Figure 9). The down confirmation is similar to the up confirmation 56 described above but omits the initial correlation sequence, which is not required on the downlink. Thus, the down confirmation contains a SEQ sequence number, and CRC bits. If the down confirmation indicates that all cells were received successfully, no further action is required. However, if the down commit specifies cells that were most likely not received, the commit tells the SU which cells it needs to be forwarded. The Ap then allocates additional bandwidth to forward those cells, using a descriptor field 76 type 10 in the upstream frame descriptor. The SU then sends the failed cells, again with linear pre-distortion. The down confirmation procedure is then repeated as required.

If the SU has requested access to send more cells than can be scheduled in a single frame, then the AP will schedule additional housings, signaled using type 10 upstream descriptor fields, as required.

Error Recovery in the Rising Cell Process

At each stage of the upstream access state machine, there is the possibility that an impulse will be lost or degraded. The system will recover via reprogramming if any pulse fails, as illustrated in Figure 25. This ensures robust communications even if individual transmitted pulses are lost or degraded.

Descending and Ascending Impulse Structures

These are as described above, but are also set by reference in Figures 26 and 27.