PROVIDING OPTIMAL UPLINK DATA RATES TO MOBILE STATIONS WHEN MOBILE NETWORKS SEND PARAMETERS CONTROLLING THE DATA RATES

The invention relates to wireless communications, and more specifically to a method and apparatus for providing optimal uplink data rates to mobile stations when mobile networks send parameters controlling the data rates. BACKGROUND

Mobile networks are often used to provide communication between mobile devices. In general, mobile networks contain systems such as base stations which communicate over air with individual mobile devices using 3G and IEEE802 cellular technologies such as CDMA20001xEV, UMTS/HSDPA/HSUPA, and IEEE802.16e/WiMAX. Additional communication infrastructure is provided to facilitate communication between distant mobile devices.

SUMMARY

An often important consideration in mobile environments based on digital communication technologies is the rate (amount of data per unit time) at which a mobile station can send data to the mobile network. The data rate is generally referred to as uplink data rate. In general, a higher, uplink data rate implies that the mobile station can transmit correspondingly more data per unit time (in the allocated time slots) to the mobile network. Uplink data rates are often controlled by parameters received from mobile networks. For example, in the CDMA 2000 environment, a base station sends parameters known as reverse activity bit (indicating whether the uplink rate can be increased or decreased), rate limit (maximum rate at which the mobile station can transmit), and transition probabilities (indicating the probability with which the data rate can be increased or decreased) etc., which are then used by each receiving mobile station to determine the corresponding uplink data rate. The manner in which the parameters control the uplink data rate is described in further detail in a standard document entitled, "CDMA 2000 high rate Packet data air interface specification", available from Third Generation Partnership Project 2 (3GPP2).

There is a general need to provide optimal uplink data rate to mobile stations. Providing too high a data rate may be considered "unfair" since the data rates of other mobile stations may be correspondingly reduced (to keep interference within manageable levels).

On the other hand, providing too small a data rate may lead to several disadvantages. For example, in situations in which several types of multi-media applications are supported in the mobile network and mobile stations, too low a bandwidth may impede the ability of mobile stations to provide desired quality of service (QoS) (e.g., low latency for voice or bursty/high transfers for data networks access applications) to different applications.

Accordingly, what is needed is a method and apparatus to provide optimal uplink data rate in environments such as those described above. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram illustrating an example environment in which several aspects of the invention can be implemented.

FIG. 2 is a flowchart illustrating the manner in which uplink data rate is updated according an aspect of the invention.

FIG. 3 is a block diagram illustrating the details of a mobile station in an embodiment of the invention. DETAILED DESCRIPTION OF THE EMBODIMENTS 1. Overview

A mobile station provided according ,to an aspect of the invention determines the uplink data rate based on present ability of the mobile station to meet the QoS requirements of individual applications, as well as the parameters (designed to control the uplink data rate) received from (and originating in) the mobile network. Due to the adjusted uplink data rate, the mobile station may be able to better meet the QoS requirements of the individual applications.

In an embodiment implemented in the context of CDMA 2000 architecture, the parameters received include reverse activity bit and rate limit noted above, and transition probabilities p and q (described below in further detail) specified for each flow supported by a mobile station. The present ability to meet the QoS requirements is based on the 1. wait time of packets in queues/flows of interest, 2. depth of queues of interest, 3. difference between the current data rate and the data rate potentially specified for individual application, and 4. number of failures to timely transmit packets in pre-specifϊed prior duration.

According to another aspect of the invention, the base station examines the queues to determine various parameters impacting the observed QoS, and dynamically adjusts the transition probabilities based on the determined parameters. In an embodiment, the parameters include delay bound requirement (maximum permissible delay in transporting the packet to the destination) of packets and/or average delay or data flow rate requirement.

Several aspects of the invention are described below with reference to examples for illustration. It should be understood that numerous specific details, relationships, and methods are set forth to provide a full understanding of the invention. One skilled in the relevant art, however, will readily recognize that the invention can be practiced without one or more of the specific details, or with other methods, etc. In other instances, well known structures or operations are not shown in detail to avoid obscuring the features of the invention. 2. Example Environment

FIG. 1 is a block diagram illustrating the details of an example mobile communications environment in which various aspects of the invention can be implemented. The environment is shown containing mobile stations HOA and HOZ, base terminal station (BTS)120, base station controller (BSC) 140, packet data serving node (PDSN) 150, packet network 170 and application server 190. Each system is described below in further detail.

At least from the description below, it may be appreciated that BSC 140 and BTS 120 form part of a mobile network, packet network 170 operates according to protocols such as Internet Protocol (IP), and that PDSN 150 provides the necessary interface between the mobile network and the packet network. Only example systems are included in the figure for illustration. Typical environments contain many more systems, for example to provide voice telephone calls to distant users over PSTN network or another mobile network. In addition, only the details of various systems as relevant to an understanding of various features of the invention are provided herein for conciseness. In an embodiment, the mobile network is implemented consistent with the CDMA 2000 architecture, the data from/to application server 190 is encoded according to IS856 standard in the mobile network, and according to Internet Protocol in packet network 170. For further details, the reader is referred to the corresponding standard documents.

Mobile stations HOA through HOZ provide wireless communications to mobile users by interfacing with the mobile network. The description below is provided with reference to mobile station 11OA for conciseness. However, the description relates to other mobile stations as well. Application server 190 provides several services, which can be accessed via packet network 170. For example, application server 190 may correspond to a web server which receives HTTP requests and provides the corresponding data in the form of web pages. Packet network 170 provides transport for various data packets, and can be implemented according to protocols such as Internet Protocol and ATM, as is well known in the relevant arts.

PDSN 150 performs the necessary protocol translations in transferring data between packet network 170 and the mobile network. For example, the data in the payload of the data packets received from packet network 170 may be encoded in radio packets consistent with the protocol requirements of the mobile network, and provided to BSC 140. Similarly, the data in the radio packets received from BSC 140 is encoded in the payload of data packets suitable for transmission on packet network 170.

BSC 140 controls the operation of several base terminal stations, though only one , base terminal station is shown for conciseness. BSC 140 provides features such as radio access (i.e., access to mobile stations) and link maintenance (soft handoff when mobile station moves from one cell to the other). In addition, the radio packets to/from application server 190 are also forwarded.

BTS 120 provides the radio link for communication with mobile stations 11OA through 11OZ. BTS 120 also sends various parameters which control the uplink data rate from each mobile station. However, various aspects of the invention adjust the uplink data rate based on other factors also, as described below in further detail. 3. Providing Optimal Uplink Data Rate

FIG. 2 is a flowchart illustrating the manner in which optimal uplink data rates may be provided to mobile stations according to an aspect of the invention. The flowchart is described with respect to FIG. 1 and CDMA 2000 architecture merely for illustration.

However, the approaches can be implemented in other environments as well. The flowchart begins in step 201, in which control passes to step 210.

In step 210, mobile station HOA receives parameters originating from the mobile network, with the parameters being designed to control uplink data rates. For example, BSC 140 in combination with BTS 120 (collectively referred to as a base station, BS) may generate the control parameters based on the overall traffic in the mobile cell, and the parameters may be received via BTS 120. In the case of CDMA 2000, the parameters include reverse activity bit, rate limit, and transition probabilities p and q which are described in a section below briefly. In step 230, mobile station 11 OA determines the uplink data rate according to the parameters, but adjusted according to the present ability to meet the QoS requirements of individual applications presently executing on the mobile station. The present ability may be determined based on present as well as past information. An example approach to such determination is described in sections below. The flowchart ends in step 299. Due to such adjustment, various benefits such as meeting effectively the QoS requirements of various multimedia applications supported by mobile station HOA may be obtained. In addition, given the shared nature of the aggregate bandwidth, higher uplink data rates may be used by mobile stations needing such higher rates due to the operation of various features of the invention. Thus, the aggregate bandwidth may be more effectively utilized.

From the above, it may be appreciated that optimal uplink data rate may be used by mobile stations due to the operation of step 230. In one embodiment implemented in the context of CDMA 2000, the uplink data rate is determined by first compensating the parameters according to the present ability to meet the QoS requirements, and then using the compensated parameters to determine the uplink data rate. Accordingly, the parameters in such an environment are described first, followed by the manner in which the parameters are compensated. 4. Parameters in CDMA 2000

Broadly, there are four relevant parameters designed to control the uplink data rate in CDMA 2000. For conciseness, only a brief description of the parameters is provided. For

further description of the parameters, the reader is referred to a document entitled, "CDMA 2000 high rate Packet data air interface specification" noted in sections above. Each parameter is described briefly.

A reverse activity (RA) bit indicates whether the total reverse traffic channel interference received in the cell exceeds a pre-specified value or not. Excessive interference generally implies that the uplink data rates need to be reduced so as to bring the interference to acceptable levels.

A rate limit indicates the maximum uplink data rate that might be used by a mobile station. Transition probabilities (p and q) are specified for each flow in a mobile station and determine the probability with which the mobile station may decrease or increase the uplink data rate respectively.

In general, these four parameters are computed by BSC 120 and BTS 140 (together referred to as a base station) and transmitted to the mobile stations via BTS 120. As noted above, according to an aspect of the invention, these parameters are compensated based on the present ability to provide desired QoS to applications locally in each mobile station. The compensation approaches generally depend on the internals of thp corresponding mobile station, and accordingly the description, is continued with respect to the details of an example mobile station. 5. Mobile Station

FIG. 3 is a block diagram illustrating the details of mobile station 11OA in one embodiment. The mobile station is shown containing applications 310A and 310B, network layer queues 320A and 320B, media access control (MAC) queues 330A and 33OB, rate determination block 350 and transceiver 370. Each block is described below in further detail. Applications 31OA and 31OB represent example applications presently executing on mobile station 11OA. For illustration, only two applications are shown, however, several more applications can be executing within mobile station 11OA. It should be understood that different applications have different desired QoS. For example, the voice/video applications require low delay bound, low jitter and minimum required data rate, while some TCP applications such as web browsing perform better with lower average delay. FTP

applications running over TCP may require some sort of throughput guarantees. In general, each application generates some network layer (e.g., IP) packets, which are to be transmitted to corresponding end system (e.g., application server 190).

Network layer queues 330A and 330B provide buffers to IP packets awaiting transmission from applications 31OA and 31OB. The packets are converted into a form (e.g., by segmentation into multiple MAC packets according to specification entitled, "TSG-C C.S0003-D vl.0" available from Third Generation Partnership Project 2 (3GPP2)) suitable for transmission on mobile network (to BTS 120) and stored in corresponding MAC layer queue. Thus, MAC packets corresponding to network layer queues 330A and 330B are respectively stored in MAC layer queues 330A and 330B.

A retransmit queue is also associated with each transmit MAC queue. The queue contains the MAC packets which need to be transmitted again, for example, because BTS 120requests for retransmission.

Transceiver 370 transmits a next MAC packet from a specific queue 330A/330B once the mobile network indicates that the mobile station can transmit in the present time slot. A mobile station may send data from a desired flow in that time slot. Multiple mobile stations may send data in uplink, in the same time slot. If the mobile network indicates tp mobile 110- A that it is allowed to send data in a specific slot, this mobile HO-A can send data from 33OA and/or 330B. The rate of transmission is determined by rate determination block 350. The manner in which rate determination block 350 determines the uplink data rate according to various aspects of the invention, is described below. However, it is helpful to first establish some terminology to describe the related approaches. 6. Terminology

It is first helpful to note that multiple mobile stations generally interact with a mobile network, and each mobile station may contain multiple flows. For example, with respect to FIG. 3, two flows are corresponding to the two applications respectively.

represents a delay (d) and jitter (jt) sensitive reverse link flow (rf) corresponding to flow m within mobile station z. abound (m,z)

represents the delay bound requirement of reverse link flow (m) of mobile station (z) over the air-interface between mobile station and BTS. This term indicates the maximum amount of time delay (from the time placed in queue 320A/320B) permissible in transmission of IP packets from mobile station 11OA to BTS and may be set by the mobile station based on the QoS requirements for this flow (or the supported application in general). An application may have an end-to-end delay requirement (from mobile station to another mobile station or an application server) and dbound(m,z) represents the delay bound requirement over the air- interface between mobile station to BTS. djitter(m,z). is the delay jitter of reverse link flow m of mobile station z. This term indicates the maximum permissible delay variation experienced by any two consecutive IP packets and may be set by the mobile station based on the QoS requirements for this flow. denotes IP packet (p) of flow rf(d,jt)[m,z].

denotes MAC packet (k) of IP packet p, wherein k can take integer values ranging from 1 to x(p) where x(p) denotes the number of MAC packets that IP packet p has been segmented into. denotes the buffer depth/length of MAC packet queue of reverse link flow m of mobile station z at the beginning of slot n.

and q(m,z)[n] represent the transition probabilities p and q received from the mobile network at time slot n. and

represent the compensated transition probabilities p and q at time slot n, determined at the mobile station.

Pbase(m,z)[ή] and represent the initial (prior to present recomputation) values of the transition probabilities for reverse-link flow m of mobile station z, that are determined in the mobile network. wait(m,z)[h,n]tepvesmts the duration of time that MAC packet h of flow m of mobile station z has been in its queue by the beginning of time slot n.

Time slot n '* prev represents a time slot prior to time slot n.

The manner in which the present ability to provide desired QoS can be determined, is described below with an example.

7. Present Ability to Provide Desired QoS

In one embodiment, the present ability to provide desired QoS is based on four measurements:

1. waiting time of packets in queues of interest;

2. Depth of queues of interest;

3. the difference between the required data rate and the current data rate and 4. Number of failures to timely transmit packets in a pre-specified prior duration. Broadly, (1) and (2) are relevant to applications which would require packets to be dropped if not transmitted by corresponding deadlines. On the other hand, (2) and (3) are more relevant in case of data applications characterized by bursty traffic, but with tolerable latencies in transmitting packets.

It should be appreciated that different combinations of above measurements may be used in different scenarios to determine the present ability to meet the QoS requirements. The manner in which the parameters (designed to control the uplink data rate) can be compensated in example scenarios, using combinations of the above measurements, is described below in detail.

8. Compensating Parameters in Scenario 1

In this scenario, mobile station 11OA dynamically updates the values of transition probabilities p and q (within certain bounds) for reverse link flows with delay bound requirements based on measurement 1, i.e., waiting time of packets in queues of interest, as described below in further detail. A single logical queue is created for MAC packets associated with a particular flow in which the MAC packets are stored in ascending order of the value of the term {dbound(m,z) - wait(m,z)[h,n]} with the head-of-line (HOL) packet in the queue being the one with the smallest value of this term. Here, dbound and wait terms are as defined in a previous section.

As can be readily seen, the term dbound(m,z)-wait(m,z)[h,n] is a measure of how close a particular MAC packet is to violating its delay bound requirement. Smaller values of this term mean that the corresponding MAC packets need to be transmitted quickly, or else dropped. Accordingly, the transition probabilities p and q for the flow are modified according to the below equations:

Wherein

and

is constrained by:

The lower and upper limits noted above can be pre-specified or computed dynamically based on the policies of service operators.

Similarly, the compensated p value may be bound according to the following equation:

Wherein γ(m,z) and can be pre-specified by an operator or computed dynamically based on policies of the service operators.

The maximum number of entries in the MAC queue (single logical queue referred to above) to be examined may be set according to the following equation:

Now the manner in which the q value can be compensated is described. The q value can be compensated according to:

Equatιon(6) wherein

is a pre-confϊgured normalizing constant (set generally determined based on the operator policies) . The remaining terms have been defined above. q _AT (m,z)[n] may be constrained accordin to:

Wherein

and can be pre-specified by an operator or computed dynamically based on policies of the service operators. The manner in which uplink data rate can be determined based on the compensated p and q values, is described below in further detail.

9. Compensating Parameters in Scenario 2

In this scenario, mobile station dynamically updates the values of transition probabilities p and q (within certain bounds ) of reverse link flows with delay bound requirements based on measurements 1 and 2, i.e., waiting time of packets queues of interest and depth of queues of interest, as described below. Transition probability p can be compensated according to the following equation:

Wherein

parameter is defined in previous section(s), and σ parameter is computed according to the following equation: , _n . {9) wherein

represents the queue depth of flow m of mobile station z at the beginning of slot n, qdepth _max(m,z) represents the maximum allowable queue depth (corresponding to flow m) of mobile station z, tf_rate(m,z) represents a pre-specified average rate at which flow m of mobile station z can send data, represents a pre-specified positive constant and depends upon the quality of service (QoS) for that flow and upon policies, such as pricing, of network operators, and norm_const also represents a pre-specified constant. Further Base station generally contains traffic specification and QoS requirements of each flow provisioned on a mobile station and thus can compute these fixed queue depths for each mobile - qdepth_max(m,z). Also, such information may be provided to base station by the mobile in the session setup signaling.

is constrained according to:

Wherein

and

are pre-specified constants (determined by operator policies).

The compensated value of p is further constrained according to:

wherein γ(m,z) and

can be pre-specified by an operator or computed dynamically based on policies of the service operators.

Similarly, parameter q is compensated according to:

wherein

and are as defined earlier.

and

are pre-specified constants chosen according to the below equations:

The compensated q may be constrained according to:

wherein and can be pre-specified or computed dynamically based on the policies of the service operators.

It should be appreciated that the approach described above can be used when compensation is sought to be attained for a combination of more than one measurement. Various combinations of measurements can be used for combination, as would be appropriate under the corresponding scenario. 10. Compensating Parameters in Scenario 3

In this scenario mobile stations are allowed to dynamically update the values of transition probabilities p and q (within certain bounds) of reverse link flows with rate requirements, based on measurement 3, i.e., the difference between the required data rate and the current data rate.

The compensated value of p is computed according to:

Wherein is computed according to:

wherein

represents the required rate for flow m of mobile station z, and currentRate represents the corresponding current/present rate.

The compensated value of p may be constrained according to: wherein the values of γ and β can be chosen such that equations 22 and 23 noted below are satisfied. The compensated value of q is computed according to:

wherein the values of

is a pre-specified normalizing constant (determined based on the operator policies) The compensated q may be constrained according to:

wherein and can be chosen such that equations 22 and 23 are satisfied.

Parameters

(in equation 14),

(in equation 15),

are chosen such that equations 22 and 23 given below are satisfied.

wherein

and

represent the transition probability parameters p and q respectively of a reverse flow m_d, which has delay bound requirements, and belongs to mobile station z_d, and and

represent the transition probability parameters p and q respectively of a reverse flow m_r, which has rate requirements, and belongs to mobile station z_r.

11. Compensating Parameters in Scenario 4

In this scenario, mobile stations are allowed to dynamically update the values of transition probabilities p and q (within certain bounds) of reverse link flows with delay bound and jitter requirements, based on measurements 1, 2 and 4, i.e., waiting time of packets in queues of interest, depths of queues of interest and the number of failures to timely transmit packets in pre-specified prior duration, as described below.

In this scenario, the transition probability p can be compensated according to the following equation: Eq(24) wherein

and are computed according to:

and

are computed as given earlier in equations 9 and 2 respectively. is constrained according to

is constrained according to

Transition probability q can be compensated according to the following equation:

wherein ,

and are normalizing constants(determined based on the operator policies).

Thus, using approaches such as those described above, the transition probabilities can be compensated. The compensated probabilities can then be used to determine the uplink data rates, as described below with an example approach.

12. Determining the uplink data rate

Mobile station HOA may determine the optimal uplink data rate based on the compensated values of transition probabilities p and q, as follows.

At the mobile station a computation of the term CombinedRAbit is done, wherein 5 CombinedRAbit is defined to be the logical OR operation of the most recent

RA bits received from all cells. Thus, if excessive interference is detected in any of the cells

(as indicated by a 1 of the corresponding RA bit), the CombinedRAbit is set to 1.

The uplink data rate can be determined based on the value of CombinedRAbit using the following approach: 0 When CombinedRAbit has a value 1, the uplink data rate can be determined using:

Note: have changed 9.6 kbps to minjrate below.

MaxRate(m, z)[ri] = max(min_ rate, CurrentRate(m, z) 12) with probability 5 or, equivalently:

MaxRatφn, z)[n] - max(min_rafe, CurrentRate(m, z)) with probability (1 - p _AT (m, z)[n]) ....εq (31) wherein: ^'

MaxRate(m, z)[n] is the maximum uplink data rate o CurrentRate(m, z) is the current data rate min_rate is the minimum non-zero data rate at which a mobile is allowed to send data to base station on reverser link. This rate is pre-specified for a wireless system.

That is, from equation 30, the newly determined value of the maximum uplink data 5 rate is the greater of half the current data rate and minjrate, where the probability of this determination is given by the value of the compensated transition probability parameter p _AT (m,z)[n]

Equivalently, as represented by equation 31, the newly determined value of the maximum uplink data rate is the greater of the current data rate and min_rate, where the probability of this determination is given by

where

is the value of the compensated transition probability parameter p

On the other hand, when CombinedRAbit has a value 0:

or, equivalentfy:

wherein: is the maximum uplink data rate CurrentRat e(m,z) is the current data rate max rate is the maximum rate at which a mobile station is allowed to send data to a base station.

That is, from equation 32, the newly determined value of the maximum uplink data rate is the lesser of twice the current data rate and 153.6 kbps, where the probability of this determination is given by the value of the compensated transition probability parameter

]

Equivalently, as represented by equation 33 A, the newly determined value of the maximum uplink data rate is the lesser of the current data rate and max_rate, where the probability of this determination is given by

where

is the value of the compensated transition probability parameter q.

Thus, the four example scenarios presented above illustrate how mobile stations in a network can adjust their uplink data rates by compensating certain parameters based on measurements which indicate present ability to provide a desired QoS. This dynamic adjustment can be made at short intervals (say, of the order of a few mille-seconds). It should be appreciated that the compensation approaches described above may cause mobile stations 11 OA-11OZ to increase the uplink data rates. To the extent such increases can be supported in the further data path, such increases may be permitted. On the other hand, if the increases cause QoS violations of various flows, it may be desirable to cause mobile stations to reduce the corresponding uplink data rates. The manner in which a base station can compute transition probabilities to support such a feature is described below. 13. Computation of Transition Probabilities in Base Stations

Base station 120 can compute transition parameters while taking into consideration some of the measurements (but performed for the queues within base station 120) noted above. For example, the computation can take into consideration the fraction of packets of a reverse-link flow (from mobile station towards the application server 190) which have experienced delay, rate or average delay violations, depending on the specific nature of the flow. The computed parameter values can then be transmitted to the mobile stations where the uplink rate determination is done according to the approaches described above.

With respect to computation of the transition probabilities for flows with delay bound requirements, parameters p and q are further recomputed if frac _ delay _viol __sys(m,z)(n) > frac _delay _viol _sys __threshl{m,z) Eq.(33B) wherein frac_delay_viol_sys_threshl(m,z) Equation(33C) is a pre-specified threshold and frac _ delay _ viol __sys(m,z)(n) Equation (33D) is computed according to equation 36 below. Here frac _ delay _ viol __ sys(m, z)(ή) is the fraction of packets for reverse link flow m of mobile station z for which delay bound was violated by the beginning of slot n, in the system, that is from the mobile station to base station

The compensated value of p is computed according to:

wherein

₅ wherein is the number of IP packets for which delay bound was violated while these packets were queued in the radio access network, _{j Q} is the number of packets of flow m of mobile station z which arrived at the PDSN, is the number of packets dropped by slot n (time slot at which computation is being . performed), norm_const_II is a normalizing constant,. and γr

The compensated value of q is computed according to:

wherein 0 is a normalizing constant.

The recomputed values of p and q are constrained to lie within some pre-specified limits.

With respect to computation of the transition probabilities for flows with rate requirements, transition probabilities p and q are recomputed if 25 wherein

is a pre-specified threshold and is computed according to equation 40 below. 5 The compensated value of p is computed according to:

wherein

_Q

is the required rate of flow m of mobile station z,

is the observed rate of flow m of mobile station z by the beginning of slot n. 2 is a normalizing constant

The compensated value of q is computed according to:

wherein Q is a normalizing constant.

The compensated values of p and q are constrained to lie within some pre-specified limits.

With respect to computation of the transition probabilities for flows with average 5 delay requirements, transition probability parameters p and q are recomputed if

wherein is a pre-specified threshold and is computed according to equation 44 below.

The compensated value of p is computed according to:

wherein

is the required average delay for flow m of mobile station z. is the observed average delay of flow m of mobile station z by the beginning of slot n, in the radio access network, and is a normalizing constant

The compensated value of q is computed according to:

wherein

is a normalizing constant

The compensated values of p and q are constrained to lie within some pre-specified limits. 14. Example Implementation of Base Station

FIG. 4 is a block diagram illustrating the details of a base station in one embodiment. The base station is shown containing base terminal station (BTS 401), base station control (BSC) center 402 and rate determination block 460. Further BTS 401 is shown containing transceiver 470 and MAC layer queues 430A and 430B; BSC 402 is shown containing network layer queues 420A and 420B and network interface 410. Each block is described below in further detail.

Transceiver 470 receives MAC packets from multiple mobile stations and sends the received MAC packets to appropriate MAC layer queue 430A or 430B. Each MAC layer queue 430A and 430B may contain a transmit queue and a receive queue. Accordingly, transceiver 470 may send MAC packets received from mobile stations to the receive queue and may transmit a MAC packets in the transmit queue received from BSC 402.

BSC 402 accesses MAC layer queue on paths 432A and 432B, and combines the data in multiple MAC packets to generate a corresponding network layer (IP) packet. The IP packets are stored in network layer queues 420A/420B. Paths 432A and 432B may represent various communication channels such as RF, cable, OFC etc. Network interface 410 forwards the IP packets in the queues to the next device (e.g., PSTN) or to an IP network, as appropriate.

Rate determination block 450 examines the data (e.g., packet headers, queue lengths) in the queues (MAC layer queues 430A and 430B, and network layer queues 420A and 420B) to determine various parameters defining the QoS requirements, and dynamically adjusts the transition probabilities based on the determined parameters. As both MAC and network layer queues are being examined, rate determination block 450 may be implemented in a distributed manner, as logically represented in the figure.

In an embodiment, the parameters include a number of packets violating a delay bound requirement (maximum permissible delay in transporting the packet to the destination), average delay being experienced in the corresponding flow in a desired duration, and the data flow rate for the corresponding flow. The equations noted above are used to determine the transition probabilities p and q. The determined probabilities are then transmitted to the mobile stations by interfacing MAC layer queues 430A and 430B.

The delay requirement may be measured using one of various techniques. For example, in TCP/IP environments, the source application devices may be configured to include a time stamp in the TCP header of each packet, and the time stamp can be used to determine the delay experienced by the packet thus far. Various GPS type technologies can be used to synchronize clocks (at the source of the packets and the base stations) to accurately determine the delay values. The PDSN may look at transport layer packet headers and send this information back to the rate determination block 450.

The average delay may also be computed from the measured delay by considering a number of packets over a desired duration. Similarly, rate determination block 450 may check the data flow rate for each flow based on the number of network / IP packets or the aggregate payload lengths being transported by the packets.

Due to the use of such techniques, the transition probabilities can also be optimally computed according to various aspects of the invention. It should also be appreciated that various features described above can be implemented in a combination of one or more of hardware, software and firmware. The description is continued with respect to an embodiment in which the features are operative upon execution of appropriate software instructions from a machine readable medium. 15. Machine Readable Medium

FIG. 5 is a block diagram of computer system 500 illustrating an example system in an embodiment of the invention. System 500 may correspond to each of a portion of a base station and a mobile station. Computer system 500 may contain one or more processors such as central processing unit (CPU) 510, random access memory (RAM) 520, secondary memory 530, graphics controller 560, display unit 570, network interface 580, and input interface 590. All the components except display unit 570 may communicate with each other over communication path 550, which may contain several buses as is well known in the relevant arts. The components of FIG. 5 are described below in further detail.

CPU 510 represents an embedded processor such as DSP (digital signal processing) processors, ARM processor etc., well known in the relevant arts, and may execute instructions stored in RAM 520 to provide several features of the invention. For example, the mobile station may examine parameters such as delay values and queue lengths, and

determine the transition probabilities based on the QOS requirement as described above. Base station may compute the transition probabilities as described above.

CPU 510 may contain multiple processing units, with each processing unit potentially being designed for a specific task. For example, a DSP processor may implement MAC layer queue 330A and 330B. An ARM processor may perform task of Network layer queue and applications. In the case of a mobile station, CPU 510 may contain only a single processing unit. RAM 520 may receive instructions from secondary memory 530 using communication path 550.

Graphics controller 560 generates display signals (e.g., in RGB format) to display unit 570 based on data/instructions received from CPU 510. Display unit 570 contains a display screen to display the images defined by the display signals. Input interface 590 may correspond to a key_board and/or mouse, and generally enables a user to provide inputs. Network interface 580 contains various antennas and other interfaces needed to communicate with external devices. Secondary memory 530 may contain hard drive 531, flash memory 536 and removable storage drive 537. Secondary storage 530 may store the software instructions and data, which e.nable computer system 500 to provide several features in accordance with the invention.

Some or all of the data and instructions may be provided on removable storage unit 540, and the data and instructions may be read and provided by removable storage drive 537 to CPU 510. Floppy drive, magnetic tape drive, CD_ROM drive, DVD Drive, Flash memory, removable memory chip (PCMCIA Card, EPROM) are examples of such removable storage drive 537.

Removable storage unit 540 may be implemented using medium and storage format compatible with removable storage drive 537 such that removable storage drive 537 can read the data and instructions. Thus, removable storage unit 540 includes a computer readable storage medium having stored therein computer software and/or data. An embodiment of the invention is implemented using software running (that is, executing) in computer system 500.

In this document, the term "computer program product" is used to generally refer to removable storage unit 540 or hard disk installed in hard drive 531. These computer program

products are means for providing software to computer system 500. As noted above, CPU 510 may retrieve the software instructions, and execute the instructions to provide various features of the invention. 16. Conclusion While various embodiments of the invention have been described above, it should be understood that they have been presented by way of example only, and not limitation. Thus, the breadth and scope of the claimed invention should not be limited by any of the above described example embodiments, but should include all such variations and other embodiments thereof as would occur to those skilled in the art to which the invention relates based on the disclosure made herein.