Closed Loop Resource Allocation in a High Speed Wireless Communications Network

ES 2 279 770 T3

DESCRIPTION

Closed-loop resource allocation in a high-speed wireless communications network.

Background of the invention

I. Field of the invention

The present invention relates to wireless communications. More particularly, the present invention relates to a new and improved method and apparatus for determining data transmission rates in a high speed wireless communication system.

II. Description of the prior art

A modern communication system is required to support various applications. One such communication system is a code division multiple access (CDMA) system that conforms to the TIA / EIA Subscriber Station-Base Station Compatibility Standard (IS-95 for Cellular System of Bandwidth Spreading). Dual Mode Wide ”hereinafter referred to as the IS-95 standard. The CDMA system enables voice and data communications between users over a terrestrial connection. The use of CDMA technique in a multiple access communication system is described in US Patent No. 4,901,307. entitled "SPREAD SPECTRUM MÚLTIPLE ACCESS COMMUNICATION SYSTEM USING SATELLITE OR TERRESTRIAL REPEATERS" and United States Patent No. 5,103,459, entitled "SYSTEM AND METHOD GENERATING WAVEFORMS IN A CDMA CELLULAR TELEPHONE SYSTEM" both transfers to the assignee of the present invention.

In this specification, "base station" refers to the hardware with which the subscriber stations communicate. Cell refers to material equipment or geographic coverage area, depending on the context, in which the term is used. A sector is a partition of a cell. Since a sector of a CDMA system has the attributes of a cell, the teachings described in terms of cells are easily extended to sectors.

In the CDMA system, communication between users is carried out by one or more base stations. A first user at a subscriber station communicates with a second user at a second subscriber station by transmitting data on the reverse link to the base station. The base station receives the data and can route the data to another base station. The data is transmitted on the uplink from the same base station, or a second base station, to the second subscriber station. The uplink refers to the transmission from the base station to a subscriber station and the reverse link refers to the transmission from the subscriber station to a base station. In IS-95 systems, the uplink and reverse link are handoffs at separate frequencies.

The subscriber station communicates with at least one base station during a communication. CDMA subscriber stations can communicate with multiple base stations at the same time during soft handoff. Soft handoff is the procedure for establishing a link with a new base station before breaking the link with the previous base station. Progressive transfer minimizes the likelihood of failed calls. The procedure and system for providing communication with a subscriber station through more than one base station during the soft handoff procedure are described in US Patent No. 5,267,261, entitled "MOBILE ASSISTED SOFT HANDOFF IN A CELLULAR TELEPHONE SYSTEM ”, transferred to the assignee of the present invention. The most progressive handoff is the procedure by which communication occurs over multiple sectors that are served by the same base station. The more progressive transfer procedure is described in detail in copending United States Patent Application Serial No. 08 / 763,498, entitled "METHOD AND APARATUS FOR PERFORMING HANDOFF BETWEEN SECTORS OF A COMMON BASE STATION," filed December 11. 1996, issued to the assignee of the present invention.

Given the increasing demand for wireless data applications, the need for highly efficient wireless data communication systems is becoming increasingly important. The IS-95 standard can transmit traffic data and voice data on the uplink and reverse links. A procedure for transmitting traffic data in fixed-dimension code channel frames is described in detail in US Patent No. 5,504,773, entitled "METHOD AND APPARATUS FOR THE FORMATTING OF DATA FOR TRANSMISION", issued to the assignee of the present invention. According to the IS-95 standard, traffic data or voice data is divided into code channel frames that are 20 ms wide with data throughputs of up to 14.4 kbps.

A system entirely dedicated to high-speed wireless communications is described in copending United States Patent Application Serial No. 08 / 963,386 (the '386 application), filed November 3, 1997, entitled, "METHOD AND APPARATUS FOR HIGHER RATE PACKET DATA TRANSMISSION ", which is granted to the assignee of the present invention. In the '386 request, the base station transmits to subscriber stations sending frames that include a pilot time between data bursts multiplexed in the frame and transmitted at a rate based on the channel information transmitted from the subscriber station to the base station. .

ES 2 279 770 T3

An important difference between voice services and data services is the fact that the latter impose strict and fixed delay requirements. Typically, the overall one-sided delay of speech frames should be less than 100 ms. Rather, the data delay can be a variable parameter used to optimize the efficiency of the data communication system. Specifically, more efficient error correction coding techniques can be used that require more significant delays than can be tolerated by voice services. An exemplary efficient coding scheme for data is described in United States Patent Application Serial No. 08 / 743,688, entitled "SOFT DECISION OUTPUT FOR DECODING CONVOLUTIONALLY ENCODED CODEWORDS" filed November 6, 1996, issued to the assignee of the present invention.

Another important difference between voice services and data services is that the latter require a fixed and common quality of service (QOS) for all users. Typically, for digital systems providing speech services, this translates into a fixed and equal transmission rate for all users and a maximum tolerable value of speech frame error rates. On the contrary, for data services, the QOS can be different from one user to another, it can be negotiated and it should be subjected to some fairness restrictions. The QOS that a data communication system provides to a subscriber is typically described by the delay, the average throughput, the blocking probability, the connection loss probability experienced during the service time.

A wireless data communication system can typically provide a range of data transmission rates on both the uplink and reverse links. Data transmission rates are transferred to various active traffic sources according to a strategy, identified as medium access control, which must justify the fact that the sources typically offer different incoming information data rates, essentially depending on the application of the data. selected data. Similarly, channel conditions and overall system load should be considered when assigning data throughput to a specific subscriber.

The multiple access control amounts to allocate the resource to active subscriber stations in the network in a way that optimizes the transactional solution between overall system throughput, QOS, and algorithm complexity. Although in the uplink the “one-to-many” nature of the transmission can be exploited to realize the optimal allocation of centralized resources in the base station, in the “many-to-one” reverse link the problem of optimization of the control strategy of Medium access is complex, and can be solved with a centralized approach at the base station, or with a distributed approach at the subscriber stations. Although many of the techniques described herein can be extended to mean access control for uplink signals, the focus of the present invention is on mean access control for the reverse link.

The information that should be used to carry out resource allocation on the reverse link resides in both the base station network and the subscriber stations. Specifically, on the side of the network resides the information pertaining to the capacity of instantaneous loading and reservation of traffic of each base station. Charging can be quantified, for example by increasing the overall received energy on the basis set by the noise power spectral density. The reserve capacity is the difference between the maximum allowable load that prevents network instability and the instantaneous load. Information about the class of terminals (for example the maximum transmission power, the transmission dimension of the buffer memory, the supported data throughput set) the channel conditions (for example signal-to-signal ratio) resides at the subscriber station. noise plus interference for all received pilot signals, transmit power), and traffic source status (e.g. buffer status, buffer overflow, throughput of traffic medium and the past, delay statistics), In principle, the information can be exchanged between the network and the subscribers, but this implies signaling over the air interface which implies a loss of resources and a delay in the Decision-making process.

A first problem is therefore to design an average access control strategy for the reverse link that optimally exploits the available information by minimizing signal messages. Likewise, it is desirable that the medium access control strategy is strong from the point of view of changes in the class of subscriber stations and in the network typology. Another fundamental problem is the allocation of resources for a subscriber station in soft handoff mode. In this case the traffic load and reserve capacity of all the base stations involved in the soft handoff (identified as base stations in the active set) must be considered, possibly again minimizing signaling in the network. Yet another fundamental problem is the protection of base stations that are not in soft handoff mode with a particular subscriber station, but are nevertheless connected to the subscriber station via an electromagnetic link with comparable path loss. to the measurements in the active set. The base stations are referred to in the present invention as the candidate set.

International Patent Application WO99 / ​​09779, entitled "A METHOD AND APPARATUS FOR REVERSE LINK RATE SCHEDULING" describes multilevel programming in a communication system capable of variable throughput transmission. Multi-level programming improves reverse link utilization and reduces transmission delay in data communication. Multi-level programming comprises base station level programming, selection level programming (system level), and network level programming. Select level programming is performed for remote stations in soft handoff mode with base stations that are controlled by the same selection programmers, and station level programming

ES 2 279 770 T3 base is implemented for remote stations that are not in soft handoff mode. The programmed base station level is carried out using the residual capacity after the higher level programming has been performed. Each schedule level can have a different schedule interval.

International Patent Application WO99 / ​​17582 describes a system to inhibit the transmission of an access request message from a mobile station when either the system is unable to guarantee access to the class of service desired by the mobile or access by the mobile station. Mobile to this class of service would increase the level of interference from the target base station or a nearby base station beyond its interference capacity. In one embodiment, each base station (BSj) broadcasts a maximum data throughput access parameter and any mobile (MSn) that receives this parameter and that seeks access to a higher data throughput refrains from transmitting an access request message. . In another embodiment, each mobile (MSm) calculates from its path loss with respect to the base station (BSj), the power level it must transmit to gain access as well as the effect of this transmission on the interference level of both the target base station (BSIj) as well as nearby base stations (NS2, BSj). If transmission at the required power level increases the interference level of any base station (BSA) beyond its maximum capacity, transmission is inhibited. The system is most effective in CDMA mobile cellular radio systems.

The present invention, described below, is a novel and efficient method and apparatus designed to address and solve the aforementioned fundamental problems for a reverse link medium access control strategy.

Summary of the invention

The present invention as set forth in the appended claims is a new and improved method and apparatus for carrying out the allocation of transmission data streams on the reverse link of a high speed wireless communication network. The present invention forms on the one hand a macro control loop with the base station network and on the other hand all the subscriber stations. Each subscriber station selects a data throughput based on the amount of data queued for transmission. This flow rate is adjusted based on the available power. This adjusted transmission throughput is then adjusted again to justify protection of the base stations in the candidate set of the subscriber station. This throughput is then adjusted according to signals indicative of the load conditions of the base stations of the active set of the subscriber station. Base stations react relative to subscriber stations by measuring their instantaneous traffic load and enhancing feedback in the form of soft busy tones. The procedure is referred to herein as Closed Loop Resource Allocation.

An objective of the present invention is to optimize the average access control of the reverse link by placing the allocation of data throughput under the control of the subscriber station which has a greater amount of information by which the throughput is determined than the elements on the network side. The subscriber has information regarding the amount of information that is queued for transmission, and the amount of transmission power available, the signal-to-noise ratios plus interference in both the active set and the links in the candidate set, all of which they are essential factors in selecting a reverse link transmission throughput. Base stations that do not have this information lack a significant amount of signaling, which is undesirable.

Another object of the present invention is to prevent a subscriber station from creating unacceptable interference to candidate base stations by their reverse link transmission, thereby enforcing protection of the candidate set.

Another object of the present invention is to allow subscriber stations to offer traffic with great impulsiveness.

Another object of the present invention is to provide fairness in resource allocation between subscriber stations taking into account the average throughput in the recent past and the possible buffer overflow condition.

Another objective of the invention is to provide efficient mean access control of the reverse link without requiring any signaling in the backhaul network, between the base station transceivers and the base station controllers, even when the subscriber station is in a handover mode. progressive. This is highly desirable because it makes resource allocation independent of network architecture and associated transmission and processing delays.

Another objective of the present invention is to minimize the necessary signaling at the air interface.

Another objective of the present invention is to avoid the waste of resources that occurs when the throughput used by the subscriber stations is less than the allocated throughput. In fact, in closed-loop resource allocation, the allocated throughput and the throughput used are always coincident.

ES 2 279 770 T3

Yet another object of the present invention is to provide soft multi-bit busy tones that indicate not only whether a base station is in an overload condition or not, but also some indication of the extent of its load.

Brief description of the drawings

The characteristics, objects and advantages of the present invention will become more evident from the detailed description made below, taken together with the drawings in which the characters of like references are correspondingly identified throughout and in which:

Figures 1A-1F are flow charts illustrating the flow rate allocation and procedure of the present invention;

Figure 2 is a diagram illustrating the basic elements of the network of the wireless communication system of the present invention;

Figures 3A-3B are block diagrams illustrating the base station of the exemplary embodiment of the present invention;

Figures 4A-4B are frame diagrams illustrating the exemplary uplink frame format of the present invention; Y

Figure 5 is a block diagram of the exemplary subscriber station of the present invention.

Detailed description of the preferred embodiments

I.- Overview of the procedure

FIG. 1A is a flow chart describing the preferred method of performing closed-loop resource allocation in accordance with the present invention. In the exemplary embodiment, the present invention is used to determine the data throughput of the reverse link transmission from a subscriber station. In block 100, the subscriber station selects a desired initial throughput (R_{stage 1}) based on the buffer state. In the exemplary example, the data throughput is determined on a per packet basis.

FIG. 1B is a flow chart describing throughput selection based on buffer state in greater detail. In block 110, the subscriber station determines the number of bytes in its transmission buffer (Q_length).

In block 112, the subscriber station determines the parameters R_min and R_max., R_min and R_max they are the minimum throughput and the maximum throughput at which the subscriber station can transmit. In the exemplary embodiment, R_max for a particular subscriber station it may optionally be established by the serving supplying base station via overhead signaling. An exemplary set of flows^® in Kbps and the corresponding packet dimensions (P_dimension (R)) in bytes of information for the flow rates is illustrated in the following table 1.

<td>Flow (R) Kb / s)</td><td> 4,8</td><td> 9,6</td><td> 19,2</td><td> 38,4</td><td> 76,8</td><td> 153,6</td><td> 307,2</td>

<td>Packet dimension (byte) (Pdimension (R))</td><td> 32</td><td> 64</td><td> 128</td><td> 256</td><td> 512</td><td> 1024</td><td> 2048</td>

In control block 114, the subscriber station determines whether the number of bytes of information in the transmission buffer is greater than the packet size for the maximum transmission throughput. In the case of exemplary numerology, the maximum throughput is 307.2 Kps and the corresponding maximum packet size is 2048 bytes. If the number of bytes of information in the transmission buffer is greater than the packet size for the maximum transmission throughput, then in block 116, the variable R_{buffer memory} is set equal to R_max. If the number of bytes of information in the transmission buffer is not greater than the packet size for the maximum transmission throughput, then in block 118 the variable R_{buffer memory} it is set to the lowest available throughput at which the entire content of the transmission memory (Qiongitud) can be transmitted in a single packet.

In block 19, the subscriber station determines the throughput of its last transmission (R_above). In the preferred embodiment, this value is stored in RAM and overwritten after each transmission. In block 120, a temporary flow rate variable R is set_{stage 1} at the minimum of the good the flow rate indicated by R_{buffer memory} or twice the flow rate Κ .-. .

ES 2 279 770 T3

In the exemplary example, the subscriber station buffer is separated into two parts. A first part includes new data for transmission and the second part includes RPL (Radio Link Protocol) data, which are packets that were previously transmitted but could be retransmitted. In the preferred embodiment, a flag F is set_memoiia Buffer when the new subscriber station data buffer is almost full. In response to the setting of the almost full buffer flag, the flow selection settings are a flow selection algorithm. In a first exemplary embodiment, the subscriber station adjusts the throughput selection algorithm to bias its transmission throughput to increasing throughput, as will be described in greater detail below. In an alternative embodiment, the subscriber station transmits at a higher predetermined throughput. It is to be understood that one skilled in the art can modify the response by setting a buffer full flag to increase throughput in various ways that are all within the scope of the present invention. For fairness reasons, the indicator F_{buffer memory} no more than N should be set_{buffer memory} times (for example 25) out of the last 100 packets.

Returning to Figure 1, the operation moves to block 102 in which the subscriber station determines the maximum throughput based on the power (R_par_{ada 2}). Figure 1C illustrating the operation performed in step 102 is in greater detail. In block 122, the subscriber station determines the power (P_max) maximum transmission rate at which the subscriber station can operate. In the exemplary embodiment, the maximum transmit power depends on the power amplifier at the subscriber station whether mobile or fixed, and on the amount of battery power at the subscriber station if the subscriber station is mobile.

In block 124, the subscriber station calculates a maximum allowed transmission power which is the maximum transmission power P_max (dN) determined in step 122 minus a power margin P_margin (dB), which allows to follow future power level fluctuations. The subscriber station then sets a variable Rp_otencia equal to the maximum throughput, R, that can be reliably transmitted with a power, p (R) (dB), less than the maximum allowed transmission power (P_max(dB) - P_margin(dB)). In block 126, the subscriber station sets a new variable R_{stage 2} equal to the minimum of R_{stage 1} determined in step 100 and R_power determined in step 124.

Returning to Figure 1A, the procedure then moves to block 104 where the subscriber station determines the maximum transmission throughput according to a protection criterion of the candidate set. The purpose of the throughput adjustment in step 104 is to protect members of the candidate set of the subscriber station from having their reverse links overloaded by subscriber stations that are not in communication with them but are sufficiently visible (from the point of view of path loss) to produce interference problems.

In the exemplary embodiment, the subscriber station is not informed of base station overload problems in the candidate set, because it does not receive the relevant busy tone. Thus, the candidate set protection algorithm is provided to prevent uncontrolled overloading of the candidate set base stations. In the exemplary embodiment, the amount of reduction in the maximum allowable transmission throughput is based on the strength of the pilot signals from the candidate base stations. In particular, the strength of the pilot signals of the candidate base stations relative to the length of the pilot signals of the base stations of the active set.

Figure 1D illustrates the exemplary procedure for determining the maximum transmission throughput based on candidate set protection. At block 128, the subscriber station measures the Ec / lo of the pilot signals from each of the base stations in its candidate set that includes all the multipath components of the base station pilots. In block 130, the base station measures the Eq / lo o f the pilot signals from each of the base stations in their active set that includes all the multipath components of the base station pilots.

In block 132, the subscriber station calculates a metric (A_ac) which is a function of the difference in strength of the signals received by the base stations in the active set and the signals received by the base stations in the candidate set. In the exemplary embodiment, the metric (A_ac) is established with respect to the difference between the sum of the EQ / lo of all members of the active set in decibels, and the sum of the EQ / lo of all the members in the candidate set in decibels, as illustrated in the following equation (1):

<img file="ES2279770T3_D0001.tif" />

where E^toc (i) / lo is the force of the nth pilot of the active set that includes all related multipath components, and E^toc (i) / l_or is the strength o f the nth pilot in the candidate set that includes all related multipath components.

ES 2 279 770 T3

In a first embodiment, the metric (A_ac) is established with respect to the difference between the weakest member of the active set and the strongest member of the candidate set as illustrated in the following equation (2):

<img file="ES2279770T3_D0002.tif" />

where E^toc (i) / l_or is the force of the nth pilot of the active set that includes all related multipath components, and E^toc (i) / l_or is the strength of the nth pilot in the candidate set that includes all related multipath components.

In a second alternative embodiment, the metric (A_ac) is established with respect to the difference between the weakest member of the active set and the sum of the members of the candidate set as illustrated in the foll owing equation (3):

<img file="ES2279770T3_D0003.tif" />

where E^toc (i) / l_or is the force of the nth pilot of the active set that includes all related multipath components, and E^toc (i) / l_or is the strength of the nth pilot in the candidate set that includes all related multipath components.

In a third alternative embodiment, the metric (A_ac) is established with respect to the difference between the strongest member of the active set and the strongest member of the candidate set as illustrated in the following equation (4):

<img file="ES2279770T3_D0004.tif" />

where E^toc (i) / lo is the force of the nth pilot of the active set that includes all related multipath components, and E^toc (i) / lo is the strength of the jnth pilot in the candidate set that includes all related multipath components.

A fourth alternative embodiment calculates the metric based on the selection of the pilot in the active set that is carrying the power control algorithm.

Other methods for determining the metric will be apparent to one of ordinary skill in the art and are within the scope of the present invention.

In block 1345, a variable R is set_dog with respect to the maximum throughput (R) so that the difference between the power required to transmit a packet from the subscriber station at the throughput R, p (R) (dB), minus a protection factor, exceeds the calculated metric value (A_ac). In the exemplary embodiment, the protection factor is determined as the power in decibels required to transmit at a flow rate that is equal to N_prot times R_min; where_prot is an integer scaling factor and R =_min number is the minimum throughput at which the subscriber station is capable of transmitting.

In block 136 a variable R_{stage 3} which is the adjusted flow rate after performing the operation of the candidate set, is determined by selecting the minimum flow rate R_{stage 2} or R_dog.

Returning to Figure 1A, at block 106, the subscriber station selects the maximum busy tone from ones received from all base stations in the active set. In a simple case, where the busy tone is a single bit indicative of the reverse link capacity loading condition or the existence of the reverse link capacity, the selection of the maximum busy tone is simply a matter of disjunction. of all received busy tones. If any of the busy tones indicates a capacity loading condition, the subscriber station stochastically reduces the throughput of its transmissions, as described below. If all busy tones indicate the additional capacity of the reverse link, then the subscriber station stochastically increases its transmission throughput as described below.

In the preferred embodiment, the busy tone is a multi-bit soft busy tone, that is, with two bits (b1, b2) corresponding to the meanings in Table 2 below.

ES 2 279 770 T3

TABLE 2

<td>(b1, b2)</td><td>Meaning</td>

<td> (0,0)</td><td>Base station lightly charged</td>

<td> (0,1)</td><td>Stable base station</td>

<td> (1,0)</td><td>Highly charged base station</td>

<td> (1,1)</td><td>Base station overloaded</td>

Figure 1E illustrates an exemplary procedure for determining two-bit busy tone values. At block 138, the base station estimates its reverse link load. There are a plurality of methods for estimating reverse link load, all of which can be applied to the present invention. The exemplary embodiment for estimating the reverse link load is described in detail in US Patent Application Serial No. 09 / 204,616, entitled "METHOD AND APPARATUS FOR LOADING ESTIMATION", which is transferred to the assignee of the present invention.

In block 140, the base station compares the estimated reverse link load with a first threshold value (TH1). If the estimated reverse link load is less than the threshold value TH1, then the base station reverse link is determined to be lightly loaded and at block 142, the busy tone bits are set to (0,0). If the estimated reverse link load is greater than or equal to TH1 then the operation moves to block 144.

In block 144, the base station compares the estimated reverse link load with a second threshold value (TH2). If the estimated reverse link load is less than the threshold value TH2, then the base station reverse link is determined to be stable and at block 146, the busy tone bits are set to (0,1). If the estimated reverse link load is greater than or equal to TH2 then the operation moves to block 148.

In block 148, the base station compares the estimated reverse link load to a third threshold value (TH3). If the estimated reverse link load is less than the threshold value TH3, then the base station reverse link is determined to be heavily loaded and at block 150, the busy tone bits are set to (1.0). If the estimated reverse link load is greater than or equal to TH3 then the operation moves to block 152. at block 152, it is determined that the base station is overloaded and the busy tones are set to (1,1).

All threshold comparisons can be performed using hysteresis loops to prevent too frequent crossovers.

At block 106, the subscriber station receives busy tones from all base stations in their active set and selects the highest busy tone.

In block 108, the transmission throughput for the current packet is selected according to the maximum busy tone (b1, b2) selected in step 106. Figure 1F illustrates the throughput selection procedure based on the selected maximum busy tone. .

In control block 154, the subscriber station determines whether the maximum busy tone (b₁, b₂₎ has the value (0, 1), which would indicate that all base stations in their active set are sparsely charged. in this case, deterministic flow increase is possible; the operation moves to control block 156, and the packet transmission rate is set to R_{stop 3} If the maximum busy tone is not set to (0,0), the operation moves to block 158.

In control block 158, the subscriber station determines whether the maximum busy tone (b₁, b₂₎ has the value (0, 0), which would indicate that at least one base station is stable (but not lightly charged). If the maximum busy tone has the value (0.1) the operation moves to the control block 160, where the stochastic flow increase is possible. In control block 160, the subscriber station determines whether the calculated throughput R_{stage 3} is less than or equal to R_previous. If R_{stage 3} is less than or equal to R_previous, then in block 162, the current packet is transmitted at rate R_{stage 3}. If R_{stage 3} is greater than R_previous, then in block 164, the current packet is transmitted at a stochastically determined rate so that the packet is transmitted at the rate R_{stage 3} with the probability po is transmitted at the flow rate R_previous with probability 1.p. If the maximum busy tone is not set to (0,1), the operation moves to control block 166.

In the exemplary embodiment, the probability (p) of increasing the transmission throughput of the subscriber station is determined based on the past activity of the subscriber station at the almost full buffer indicator (F_{buffer memory}). In particular, in the exemplary embodiment, the probability is determined based on the average flow rate in a

ES 2 279 770 T3 predetermined number of previous packets, Rmedium. In the exemplary embodiment, the probability is determined according to the equation

<img file="ES2279770T3_D0005.tif" />

if F buffer is the buffer full flag which in the exemplary embodiment assumes a value of zero or one where the buffer full condition is indicated, R_max as described above is the maximum transmission throughput of the subscriber station, N_{flow rates} is the number of streams available to the subscriber station.

In control block 166, the subscriber station determines whether the maximum busy tone (b₁, b₂) has the value (1,0) which would indicate that at least one would indicate that at least one base station in its active set is heavily loaded. If the maximum busy tone has the value (1,0) the operation moves to the control block 168, in which the stochastic flow reduction is necessary. In the control block 168, the subscriber station determines if the calculated flow rate R_{stage 3} is less than R_previous. If R_{stage 3} is less than R_previous, then in block 170, the current packet is transmitted at rate R_{stage 3}. If R_{stage 3} is greater than or equal to R_previous, then in block 172, the current packet is transmitted at a stochastically determined rate so that the packet is transmitted at the previous rate Ra with the probability po is transmitted at the higher of R_previous/ 2 or R_min with probability 1-p. In the exemplary embodiment, the number p is calculated again according to equation (5).

If the maximum busy tone does not have the value (1.0), the operation moves to control block 176 indicating that at least one base station in the active set of the subscriber station is in an overload condition. In block 176, it is determined that the transmission throughput of the current packet is the greater of Ranterior / 2 or Rmin.

II. Network description

Referring to the figures, Figure 2 depicts the exemplary data communication system of the present invention comprising multiple cells 200a-200f. each cell 200 provides service via a corresponding base station 3202 or base station 204. Base stations 202 are base stations that are in active communication with subscriber station 206 and are said to constitute the active set of subscriber station 206. Base stations 204 are not in communication with subscriber station 206 but have signals with sufficient strength to be monitored by subscriber station 206 for addition to the active set if the strength of the received signals increases due to a change in signals. propagation path characteristics. Base stations 204 are said to constitute the candidate set for subscriber station 206.

In the exemplary embodiment, subscriber station 206 receives data information from at most one base station 202 on the reverse link to each time slot, but receives busy tone information from all base stations in the active set. Likewise, the subscriber station communicates with all base stations in active set 202 over the reverse link. If the number of active base stations is greater than one, the subscriber station 206 is in soft handoff mode. Subscriber stations 202, especially those located near a cell boundary, can receive the pilot signals from multiple base stations 204 in the candidate set. If the pilot signal is above a predetermined threshold, subscriber station 206 may require that base station 204 be added to the active set of subscriber station 206. In the exemplary embodiment, before candidate base station 204 is added to the active set, there is typically no way for the subscriber station to monitor its busy tone. If a way is provided to monitor the busy tone of a candidate base station, then this busy tone enters the set within which a maximum is selected according to step 106 described above.

III. Uplink structure

A block diagram of the exemplary uplink architecture of the present invention is shown in Figure 3A. The data is divided into data packets and provided to the CRC encoder 312. For each data packet, the CRC encoder 312 generates frame check bits (eg the CRC parity bits) and inserts the tail code bit. The packet formatted from the CRC encoder 312 comprises the data, frame check bits and tail code bits, and other head bits described later. The formatted packet is provided to encoder 314 which, in the exemplary embodiment, encodes the data according to a convolutional or turbo encoding format. The encoded packet from encoder 314 is provided to interleaver 316 which rearranges the code symbols in the packet. The interlaced packet is provided to weft piercing element 318 which removes a fraction of the packet in the manner described below. The punctured packet is provided to multiplier 320 which encrypts the data with the cipher sequence from cipher 322. The output of multiplier 320 comprises the ciphered packet.

ES 2 279 770 T3

The encrypted packet is provided to the variable rate controller 330 which demultiplexes the packet into K parallel channels in phase and quadrature, where k depends on the data throughput. In the exemplary embodiment, the encrypted packet is first demultiplexed into in-phase (I) and quadrature (Q) current. In the exemplary embodiment, stream I comprises regulatory index symbols and stream Q comprises irregular indexed symbols.

Each stream is furthermore demultiplexed into K parallel channels so that the symbol throughput of each channel is fixed for all data streams. The K channels of each stream are provided to the Walsh cover element 332 which covers each channel with a Walsh function to provide an orthogonal channel. The orthogonal channel data is provided to gain element 334 which scales the data to maintain a constant total energy per chunk (and thus a constant output power) for all data streams. Data scaled from gain element 334 is provided to multiplexer (MUX) 360 which multiplexes the data with a preamble sequence. The output of the MUX 360 is provided to the multiplexer (MUX) 362 which multiplexes the traffic data, power control bits, and pilot data. The MUX output 362 comprises the Walsh I channels and the Walsh Q channels.

The reverse link power control (RPC) bits are provided to symbol repeater 350 which repeats each RPC bit a predetermined number of times. The repeated RPC bits are provided to the Walsh cover element 352 which covers the bits with the Walsh covers that correspond to the RPC indices. The covered bits are provided to gain element 354 which scales the bits prior to modulation to thereby maintain a constant total transmit power.

In addition, upstream activity is provided to the symbol repeater 350. The uplink activity bit alerts subscriber station 206 of an upcoming blank frame in which the base station will not transmit uplink data. This transmission is done to allow the subscriber station 206 to make a better estimate of the C / I of the signal from the base stations 202. The repeated versions of the upstream activity bit are Walsh covered in the Walsh cover element 352 to thereby be orthogonal to the Walsh covered power control bits. The covered bits are provided to gain element 354 which scales the bits prior to modulation thereby maintaining a constant total transmit power.

In addition, a busy tone is provided to the symbol repeater 350. The busy tone alerts subscriber station 206 of a reverse link loading condition. In an exemplary embodiment, the busy tone is a two-bit signal indicative of a request by base stations 202 for subscriber stations 206 in their coverage area to either deterministically increase or decrease the throughput of their reverse link transmissions. , or to stochastically increase or decrease the throughput of reverse link transmissions. Repeated versions of the busy tone are from Walsh covered in Walsh coverage element 352. To thereby be orthogonal to the Walsh covered power control bits and the up activity bit. The covered bit is provided to gain element 354 which scales the bits prior to modulation thereby maintaining a constant total transmit power.

The pilot data comprises a sequence of all zeros (or all ones) that is provided to multiplier 356. multiplier 356 covers the pilot data with the Walsh code W₀. Since the Walsh code W₀ is a sequence of all zeros, the output of multiplier 356 is the data. The pilot data is the time multiplexed by MUX 362 and provided to the Walsh channel I that is broadcast by the long code PN within the complex multiplier 366 (see FIG. 3B). In the exemplary embodiment, the pilot data is not broadcast with the long code PN, which is closed during the pilot burst on MUX 376, to allow reception by all subscriber stations 376. The pilot signal is thus an unmodulated PBSK signal.

A block diagram of the exemplary modulator used to modulate the data is illustrated in Figure 3B. Walsh I channels and Walsh Q channels are provided to adders 364a and 364b, respectively, which sum the Walsh k channels to provide the I signals._sum and Q_sum, respectively. Signs I_sum and Qsuma are provided to complex multiplier 366. Complex multiplier 366 also receives PN I signals and PN a Q signals from multipliers 378a and 378b, respectively, and multiplies the two complex inputs according to the following equation:

= ft ,. · W-1 - HE · • PN.Q + Q ^ 'Pffj),⁽⁶⁾ where i_mult and Q_mult are the outputs of the complex multiplier 366 and j is the complex representation. Signs I_mult and Q_mult they are provided to filters 368a and 368b, respectively, which filter the signals. The signals filtered by filters 368a and 368b are provided to multipliers 370a and 370b, respectively, which multiply the signals with COS (W_ct) sinusoidal in phase and SIN (W_ct) quadrature sinusoidal, respectively. The I modulated signals and Q modulated signals are provided to adders 372 which sum the signals to provide the modulated up waveform S (t).

ES 2 279 770 T3

In the exemplary embodiment, the data packet is broadcast with the PN long code and PN short codes. The long code PN encrypts the packet so that only the subscriber station 206 for which the packet is intended can decrypt the packet. In the exemplary embodiment, the pilot and power control bits and the control channel packet are broadcast with the PN short codes but not the PN long code to allow all subscriber stations 206 to receive these bits.

The long sequence PN is generated by the long code generator 374 and is provided to the multiplexer (MUX) 376. The long mask PN determines the deviation of the long sequence PN and is assigned only to the destination subscriber station 106. The output of the MUX 376 is the long sequence PN during the data part of the transmission and otherwise zero (eg during the pilot and power control part). The long controlled PN sequence of the MUX 376 and the short PN sequences₁ and PN_what from the short code generator 380 are provided to multipliers 378a and 378b, respectively, which multiply the two sets of sequences to form signals PN_1 and PN_Q, respectively. The PN_I and PN_Q signals are provided to complex multiplier 366.

The exemplary traffic channel block diagram shown in Figures 3A and 3B is one of numerous architectures that support data encoding and modulation on the reverse link. Other architectures, such as the architecture for the uplink traffic channel in the CDMA system that conforms to the IS-95 standard, can also be used and are within the scope of the present invention.

IV. Uplink frame structure

A diagram of the exemplary uplink frame structure of the present invention is illustrated in Figure 4A. The traffic channel transmission is divided into frames which, in the exemplary embodiment, are defined as the length of the short PN sequences or 26.67 ms. Each frame may carry control channel information addressed to all subscriber stations 205 (control channel frame), traffic data addressed to a particular subscriber station 206 (traffic frame), or may be empty (idle frame). . The content of each frame is determined by programming performed by transmitting base station 202. In the exemplary embodiment, each frame comprises 16 time slots, each time slot having a duration of 1.667 ms. A time slot of 1.667 ms is adequate to allow a subscriber station 206 to perform the C / l measurement of the uplink signal. A 1,667 ms time slot can also represent a sufficient amount of time for efficient packet data transmission.

In the exemplary embodiment, each uplink data packet comprises 1024 or 2048 bits. Thus, the number of time slots required to transmit each data packet depends on the data throughput and varies between 16 time slots for a throughput of 38.4 Kbps and 1 time slot for a throughput of 1.2288 Mbps. .

An exemplary diagram of the uplink slot structure of the present invention is shown in Figure 4B. In the exemplary embodiment, each slot comprises three of the four multiplexed time channels, the traffic channel, the control channel, the pilot channel, and the head control channel. In the exemplary embodiment the pilot signal is transmitted in two bursts and the head control channel is transmitted on either side of the second pilot burst. Traffic data is carried in the three parts of the slot (402a, 402b, 402c).

The first pilot burst 406a is time multiplexed in the first half of the slot by multiplexer 362. The second pilot burst 406b is time multiplexed in the second half of the slot. In either part of the second pilot burst 406b, the head channel data 408 including the up activity bit, busy tones, and power control bits are slot multiplexed.

In the exemplary embodiment, the busy tone is a two-bit signal and the busy tone is only set once per frame. In the exemplary embodiment, the busy tone is interleaved between the slots of a frame so that the regular slots carry the first bit of the busy tone and the irregular slots carry the second bit of the busy tone. Other ways to interleave the busy tone bits are apparent to one of ordinary skill in the art and are within the scope of the present invention.

V. Subscriber station architecture

Figure 5 illustrates the exemplary subscriber station of the present invention. Buffer 524 provides a signal indicative of the amount of data queued for transmission to throughput allocation control processor 522. The flow allocation control processor 522 selects the flow rate based on the buffer state as described in step 100 above. In the exemplary embodiment, buffer 524 is divided into two parts. A first part of the buffer 524 stores new data for transmission. A second part of buffer 524 stores data for retransmission. In the exemplary embodiment, the flow control processor 522 selects the flow rate based on a buffer full flag that is set according to the new data to be transmitted.

Transmitter 528 is responsible for upconverting, filtering, and expanding the uplink signal for transmission. Transmitter 528 provides a signal to throughput allocation control processor 522 indicative of the amount of power available for transmission of the current data packet. In response to this signal the

Flow allocation control processor 522 determines the setting to the transmit rate of the next packet as described in block 102 above.

The uplink signals are received by subscriber station 206 at antenna 500 and provided through duplexer 502 to receiver 504. Receiver 504 downconverts, filters, and expands the received signal and provides the signal to calculator 506 from pilot energy. The pilot energy calculator 506 calculates the energy of the pilot signals received from the base stations 202 of the active set and the base stations 204 of the candidate set.

The received signals are provided to the de-diffusion device 510, which de-broadcasts the pilot signals in accordance with the pilot signals from the search controller 508. In the exemplary embodiment, paging controller 508 provides a PN bypass from a candidate set or active set base station to pilot de-broadcast device 510 which in response de-broadcasts the pilot signal from a candidate set base station 204 or an active set base station 206.

The de-diffused pilot symbols are provided to the square elements 512 which calculate the symbol energy and provide the symbol energy values ​​to the accumulator 514. The accumulator 514 accumulates the energies throughout the time interval of the pilot burst and provides the pilot burst energy to flow allocation element 522. In response to the base station pilot burst energies (Ec / lo) in the candidate set and the base station pilot burst energies (Ea / lo), the flow allocation control processor 522 calculates the protection setting. of the candidate set with respect to the selected flow rate as described with respect to block 104 above.

The received signals are also provided to the busy tone demodulators 516. The busy tone demodulators 516 demodulate the busy tone values ​​for each active set base station 202 and provide the busy tone values ​​for each base station to the throughput allocation control processor 522. In response the throughput allocation control processor 522 selects the maximum busy tone as described above at 106, and calculates the transmission throughput as described above under 108.

Once the transmission throughput has been determined by throughput allocation control processor 522, a signal indicative of the selected throughput is provided to buffer 524, modulator 526 and 528. Buffer 524 sends a block of data depending on the selected transmission rate to modulator 526, modulator 526 modulates the signal according to the selected data rate and provides the modulated data to transmitter 528. The transmitter amplifies the signal according to the selected transmission throughput and provides the signal through duplexer 502 for transmission through antenna 500. The selected throughput can be indicated to active base stations through a reverse link message.

The preceding description of the preferred embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments.