Title of Invention

METHOD FOR LOAD ADAPTIVE TIMING ALIGNMENT AND UPLINK RANDOM ACCESS

Abstract The present invention relates to the field of mobile communication (3GPP L TE multi-carrier system) and particularly to the allocation of a communication resource between user equipment and base station as per the traffic load in the OFDM system. The invention describes a system for allocating the frequency-time resources between uplink timing alignment and uplink random access, dynamically based on the ratio of number of user equipments in uplink dormant mode to the number of user equipments intending to perform random access. A definite percentage will always be there for uplink timing alignment. Three methods are also suggested to reduce the amount of data being communicated to user equipments regarding the availability of random access slots.
Full Text FIELD OF THE INVENTION
The present invention, in general, relates to the field mobile communication. The invention is in the area of obtaining uplink timing alignment in a multi-carrier mobile communication. More particularly the present invention relates to system and method for load adaptive timing alignment and uplink random access.
DESCRIPTION OF RELATED ART
3GPP TS 03.64 describes a static method of obtaining timing alignment, where in the number of slots allocated for Uplink Timing alignment is reserved. The same method cannot be used in case of 3GPP LTE multi-carrier system since it would lead to wastage of uplink resources. In the proposed method UEs are allocated scheduled UL slots for timing alignment dynamically based on uplink traffic load as UEs enter and leave uplink dormant mode i.e. not having uplink data transfer on going. 3GPP LTE currently discusses the contention based uplink timing alignment on a Random access channel. 3GPP LTE does not discuss about scheduled uplink timing synchronization.
In the static method defined in 3GPP TS 03.64 for obtaining timing alignment the number of slots usable for UL timing alignment cannot be increased or decreased based on UL traffic load. The static allocation for UL timing alignment procedure is an over kill during low traffic and is not scalable for higher traffic. Also different UEs need different periodicity of timing alignment updates depending on their speeds. UEs with different periodicity requirement for timing alignment cannot be supported by the method adopted in 3GPP TS 03.64.
The contention based scheme currently being discussed by 3GPP LTE suffer -from the following disadvantages:-
1. There is high probability of collisions as the Timing alignment procedure is done on a random access channel along with uplink bandwidth request.
2. UEs have to try multiple attempts to successfully get UL scheduling request or timing alignment leading to latency in bandwidth allocation or uplink timing synchronization.
3. Due to multiple uplink attempts (due to collisions) UEs spend lot of power thereby reducing standby time.
4. Different UEs need different periodicity of synchronization depending on their distance and relative speeds with respect to base station. Both base stations and UEs can take advantage of this and do smart scheduling to optimize UE power consumption and uplink bandwidth.
5. There is no method of prioritizing between uplink timing synchronization procedure and uplink bandwidth allocation procedure.
6. UE needs to communicate UE identification and cause for Random access explicitly or implicitly for Identification by the Base station. This will reduce the number of random access opportunities on the uplink.
US patent publication bearing no. US20050130665A1 describes a system for dealing with the instantaneous load by allocating the traffic segments to the additional call attempts. The channel resources are divided into assignment and traffic segments. With each assignment segment, some traffic segments are associated, which get activated in the situation of instantaneous traffic load. The system is virtually dynamic as the segments are fixed but as situation requires, the traffic segments get attached with the assignment segments and increase their capacity. However, the publication does not describe the concept of collision reduction and synchronization with the periodicity of user equipments.
Another US patent publication no. US07027415B1 describes a method for dynamically allocating and de-allocating the communication streams based on the traffic and the traffic capacity of the already opened streams, and thus providing the flexible bandwidth to the communication system. However, the method does not mention about interferences or collisions in the system.
Another US patent publication no. US5680398A describes a system having a central unit, a remote unit, and uplink and downlink channels. To decrease the collision events, a signalling channel is used for information exchange and determining which uplink traffic channels are to be used for random access. The channels for up linking and down linking are fixed and performance is enhanced by reducing the no. of collisions. The publication does not specify the concept of dealing with user equipments of variable periodicity and making available the random access slot information to the equipments.
SUMMARY OF THE INVENTION
The invention is in the field of mobile communication where uplink timing alignment in dormant state is attained by transmitting an uplink burst and the base station calculates the time difference between actual frame beginning and actual burst arrival. In the existing timing alignment procedure, the number of uplink access opportunities is fixed and can't be changed based on instantaneous traffic/load in an adaptive manner. Different mobiles with different periodicity requirement for uplink synchronization cannot be accommodated hence forcing all mobiles to transmit at a fixed periodicity even though they could transmit with longer periodicity. This impacts power consumption for the mobile. This leads to inefficient bandwidth utilization. Also the system is not scalable as it cannot accommodate more users as the load increases.
Out of the total number of frequency - time resources allocated for Random access, a percentage is reserved for timing alignment procedure. This percentage can be varied dynamically depending on the instantaneous traffic load/random accesses on the base station. UEs in uplink dormant mode will be allocated a specific frequency - time resource and a preamble signature for requesting timing alignment. UEs send their respective signatures with/without downlink CQI (Channel Quality Indicator) or any other relevant information on their assigned uplink resources and base stations send updated timing advance values on the downlink. UEs who want to request uplink bandwidth shall do so through contention based slots which are not allocated for UL timing alignment, randomly selecting one of the available preamble sequences and frequency time resource not reserved for UL timing alignment procedure.
Due to dynamic sharing of uplink resources between uplink timing alignment procedure and uplink random access procedure, network can achieve optimal bandwidth efficiently by changing the percentage of allocation of resources to each procedure based on instantaneous traffic load on the base station. Since all UEs in dormant mode are using a scheduled channel, there are lesser collisions for bandwidth requests on contention based channel. Different periodicity of uplink timing alignment request can be supported. Since it is a scheduled transmission, base station already knows which UE sent the request and hence UEs need not carry any information bits (implicitly or explicitly) to indicate UE identification or cause of Uplink access. Due to reduced information bits carried by uplink transmission, system can support more uplink opportunities in a given frequency - time resource.
Accordingly the invention explains a method for load adaptive timing alignment and uplink random access wherein the said method involves step of allocating the frequency-time resources between uplink timing alignment and uplink random access, dynamically based on the ratio of number of user equipments in uplink dormant mode to the number of user equipments intending to perform random access.
Accordingly the invention explains a system for load adaptive timing alignment and uplink random access wherein the said system involves in allocating the frequency-time resources between uplink timing alignment and uplink random access, dynamically based on the ratio of number of user equipments in uplink dormant mode to the number of user equipments intending to perform random access.
Other objects, features, and advantages of the present invention will become more apparent from the ensuing detailed description of the invention, taken in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS
ftgure 1 depicts RACH frame structure assumed.
Figure 2 depicts Distribution of Uplink access slots
Figure 3 depicts Allocation and maintenance of UL Timing alignment slots.
Figure 4 depicts Plot of probability of collision.
Figure 5 depicts Plot of successful attempts.
DETAILED DESCRIPTION OF THE INVENTION
The preferred embodiments of the present invention will now be explained with reference to the accompanying drawings. However, it should be understood that the disclosed embodiments are merely exemplary of the invention, which may be embodied in various forms. Therefore the details disclosed herein are not to be interpreted as limiting but merely as the basis for the claims and as a basis for teaching one skilled in the art how to make or use the invention.
The present invention proposes an efficient method of attaining uplink timing alignment in an OFDM (Orthogonal frequency division Multiplexed) system by adaptive assignment of uplink resources based on instantaneous load/traffic on the cell. The total number of uplink resources are distributed among the two causes namely uplink timing alignment and uplink random access dynamically based on the ratio of number of user equipments (UEs) in uplink dormant mode to the number of UEs intending to perform random access.
Due to load adaptive distribution of uplink resources as proposed in the invention, the system is scalable and can accommodate UEs with varying periodicity requirement of uplink timing alignment and hence UEs with longer periodicity requirement can be better accommodated and their power consumption reduced.
Figure 1 depicts a frame structure based on which the proposed idea is suggested. However the proposed idea can be extended to any multicarrier system. One sub-frame is reserved in the frame of 10 ms for UL Random access or UL timing alignment purpose. RA BW as shown in Figure 1 is the bandwidth allocated for this purpose, which is 1.25 MHz. Four such frequency grids are possible in a 5 MHz band and the RA slots are frequency hopped. The sub carrier allocations are localized within an RA BW. Each Sub frame or slot consists of 75 sub-carriers (1.25 MHz) and 7 OFDM symbols (One UL frequency - time resource). 16 different preamble signatures are considered from which a UE can select one signature at random for each random-access attempt. It is assumed that, if multiple UEs make a random-access attempt in the same random-access slot and frequency band but with different preamble signatures, the network can resolve the different random-access attempts (a low preamble cross correlation). Thus, a collision is assumed to occur only if multiple UEs make a random access in the same random-access slot and frequency band using the same preamble signature.
The following aspects are to be noted:
1. Total number of frequency - time resources available for UL access (NUL-RA)
2. Ratio of UL sub-frames divided between UL timing alignment requests and uplink bandwidth request or RACH (RTA-RA)
3. RTA-RA is divided on a sub-frame basis
4. The number of UEs in dormant mode at any time is represented by (NTA)
5. The maximum UEs in dormant mode is watermarked and this water marked value can be changed depending on time averaged peak load (NTA-MAX)
6. The Number of UEs requiring UL random access (NRA)
7. The maximum Periodicity of UL Timing alignment procedure is represented as (TJA-MAX)
8. The minimum periodicity of UL timing alignment procedure is (TTA-MIN)
9. Number of preamble signatures Npreambie
Depending on the ratio of number of users in dormant mode to number of users in active mode, out of the total number of sub-frames available for UL random access (NUL-RA), A percentage of resources on a sub-frame basis are reserved for UL timing alignment procedure (NTA). This percentage can be varied dynamically depending on the instantaneous traffic load/random accesses and number of UEs in dormant mode.
UEs in uplink dormant mode will be allocated a specific sub-frame and a preamble signature for requesting timing alignment. UEs send their respective signatures with/without downlink CQI (Channel Quality Indicator) in case frequency hopping is not used on the DL, or any other relevant information on their assigned uplink resources and in response, base stations send updated timing advance values on the downlink. UEs who want to request uplink bandwidth shall do so through the usual contention based channel, by randomly selecting one of the available sub- frames not reserved for UL timing alignment procedure and randomly select a preamble sequence for transmission. Once UE is moved to idle mode or a dedicated UL resource is assigned, the scheduled UL timing alignment sub-frame is re-assigned to another UE who might be entering the UL dormant mode.
As shown in Figure 2 the sub-frames reserved for UL timing alignment should be uniformly distributed in time along with UL random access slots. The sub-frames for timing alignment and random access are located uniformly spaced in time as depicted in the lower part of Figure 2 and they are mapped to give a logical view in the upper part. In the logical view an estimate of the ratio of number of slots allocated to timing alignment and random access procedures, number of dormant mode users, the water marked maximum number of users, the number of slots available for contention based random access are shown.
It is shown in 3GPP LTE contribution R1-061228 that a UE moving at 500 Kmph would lose synchronization by a maximum of 1 micro second in 0.54 seconds. Going by this and assuming that the worst case speed is not more than 500 Kmph, the least UL timing alignment periodicity (TTA-MIN) can be set as 0.5 seconds just as an example. The periodicity of UL timing alignment procedure for UE&* could vary from TTA-MIN seconds to TTA-MAX seconds in steps of TTA-MIN seconds depending on the speed and distance of individual UEs with respect to the cell. As shown in Figure 3 the duration between any two successive UL timing alignment slots allocated to a UE should be at least TXA-MIN seconds. To support allocation of UEs with different periodicity, a bit field is to be maintained with each bit representing a timing alignment opportunity. The length of the bit field is (NPREAMBIE * Number of UL access slots in TTA-MAX seconds). When a UE enters dormant mode, it is allocated a UL timing alignment slot and a preamble for requesting updated timing alignment such that no other UE access the base station during that slot on that preamble, with periodicity (Less than TTA.MAX seconds) decided by base station or UE or mutual. The bits corresponding to the slot and preamble is marked USED for the all repetition (Considering periodicity of
UL timing alignment for the UE) within the TTA-MAX seconds such that no another UE is allocated the same slot. Once the UE moves to idle mode or to dedicated mode all these bits will be reset to UNUSED. On increased number of UL dormant users in the system, more slots can be allocated to UL timing alignment procedure on a sub-frame basis. Figure 3 shows already allocated slots for timing alignment, currently allocated slot (Yellow block) and how the slots can grow i.e. in what sequence more slots can be added (dotted blocks).
Communication of RTA-RA to UE
Since the number of reserved slots for UL timing alignment can increase or decrease with number of users in dormant mode, the number of UL access slots available for random access continuously keeps changing. Hence the number of * UL slots for random access has to be frequently broadcasted in -the downlink common control channel. The frequency of broadcast of this information decides the frequency with which RTA - RA is changed. To reduce the amount of data communicated to the UE regarding available RACH slots, two methods are suggested below.
METHOD 1
As shown in Figure 3 the allocation of slots for purpose of timing alignment shall be done in a uniform basis each timing alignment slot allocated in a periodicity of TTA-MIN- After the first slots of each TTA-MIN grid is filled, base station can start allocation from the second slot as the number of users in dormant mode increase. This is similar to water filling where the first available slots in TTA-MIN grids are
allocated. In this case the ratio can clearly identify the number and location of slots allocated for Timing alignment and random access. Hence by just broadcasting RTA - RA on the broadcast channel to UEs, the entire map can be communicated. The UE on receipt of this ratio can calculate the number and location of RACH slots available to the UE to transmit its random access burst.
METHOD 2
The distribution of total UL access slots among the timing alignment procedure and random access can be done in fixed patterns known prior to UE and base station. In this case base station needs to just communicate the pattern id currently being used for RACH slot identification.
METHOD 3
This method is a combination of the first 2 methods described above where the water filling is done as in method 1 but is done only for certain values of RTA-RA. Since it is done for only certain values of RTA-RA UE can store only these values. This method reduces the computation complexity compared to method 1 while also can save on the memory required to store all possible patterns (As some or all patterns can be realized by computation, if memory constraints don't permit storage of all patterns).
Example illustration
In this example a scheme of contention based UL timing alignment and contention based UL random access is compared with a scheme of scheduled UL timing alignment and contention based UL random access. The number of UL access opportunities available is (NPreambie * NUL-RA). In the example a large number of UEs is assumed in the cell. This implies that the collision probability can be
expressed as P = \-e'^N where N is the total number of random-access opportunities per second is taken to be 1000 and R is the random-access intensity, i.e. there are, on average, R random-access attempts per second and cell.
In Figure 4 the probability of collision is plotted as a function of R (the random- access intensity). The blue curve represents the probability of collision for a contention based UL timing alignment and contention based Random access. The green curve represents the probability of collision for Scheduled UL timing alignment and contention based random access. A can be interpreted by the plots in Figure 4 clearly the proposed scheme has a lower collision probability than the full contention based scheme.
In Figure 5 the number of successful attempts is plotted as a function of number of UL access attempts. Clearly the green curve representing number of successful attempts using a scheduled timing alignment and contention based UL random access fares better than the blue curve representing the number of successful attempts for contention based timing alignment and UL random access.
For the purpose of simulation, for the scheduled scheme, all UEs are considered to have the same periodicity of UL timing alignment and the ratio of number of users in UL dormant mode to those requesting random access is assumed to be 1:1. The UL access opportunities for UL timing alignment are allocated on a preamble + sub-frame basis where preambles are also considered as access opportunities and not on a sub-frame only basis as proposed in the invention.
ADVANTAGES
1) The probability of collisions decreases by employing scheduled UL timing alignment and contention based UL random access.
2) Due to reduced collisions, UEs need lesser attempts to reach the network there by saving lot of UE power.
3) Due to lesser attempts to reach network, the bandwidth allocation latency is reduced.
4) Due to reduced collisions there is lesser interference caused in the over all system.
5) UEs with varying periodicity for UL timing alignment can be accommodated in the proposed scheme there by not forcing UEs with lesser speeds to transmit UL timing alignment bursts more frequently than required.
6) The base station can prioritize between the UL timing alignment procedure and UL random access procedure to achieve optimal system performance.
7) Due to scheduled access, UE need not communicate UE Identification and ACCESS CAUSE to the base station. Due to reduced number of information bits carrier, more number of UL timing alignment opportunities are created.
8) Base station can do dynamic allocation of UL access slots based on number of users doing random access and number of users doing timing alignment in dormant mode based on relative priority between the two procedures. This makes the system scalable.
9) Due to watermarked upper limit on the number of access opportunities for timing alignment, the system can be immune to sudden increase in number of users in dormant mode and also be bandwidth efficient.
10) Three methods are proposed for communicating the number and location of RACH slots being available for UL random access, in which very less information bits are needed and hence very efficient in communicating to the UE on broadcast channel.
Although, the present invention has been fully described in connection with the preferred embodiments thereof with reference to the accompanying drawings, it is to be noted that various changes and modifications are possible and are apparent to those skilled in the art. Such changes and modifications are to be understood as included within the scope of the present invention as defined by the appended claims unless they depart there from.
GLOSSARY OF TERMS AND THEIR DEFINITIONS THEREOF
UEs User equipments CQI Channel Quality Indicator UL Uplink
DL Downlink
NRA Number of UEs requiring UL random access
NTA Number of UEs in Dormant mode
NTA-MAX Maximum number of UEs in dormant mode
NUL-RA Total number of frequency - time resources available
RTA-RA Ratio of UL frequency - time resources divided between UL timing
alignment requests and uplink bandwidth request
Npreamble Number of preamble signatures
OFDM Orthogonal frequency division Multiplexing
TTA-MAX Maximum Periodicity of UL Timing alignment procedure
TTA-MIN Minimum Periodicity of UL Timing alignment procedure
RACH Random Access channel





We Claim:
1. A method for load adaptive timing alignment and uplink random access wherein the said method involves the step of allocating the frequency-time resources between uplink timing alignment and uplink random access, dynamically based on the ratio of number of user equipments in uplink dormant mode to the number of user equipments intending to perform random access.
2. A method as claimed in claim 1 wherein the total number of uplink resources are distributed among the uplink timing alignment and uplink random access dynamically based on the ratio of number of user equipments (UEs) in uplink dormant mode to the number of UEs intending to perform random access.
3. A method as claimed in claim 1 wherein due to load adaptive distribution of uplink resources the method is scalable and can accommodate UEs with varying periodicity requirement of uplink timing alignment.
4. A method as claimed in claim 1 wherein, if multiple UEs make a random-access attempt in the same random-access slot and frequency band but with different preamble signatures, the network resolves the different random-access attempts.
5. A method as claimed in claim 1 wherein if the total number of frequency time resources available for UL access is given by (NUL-RA) and ratio of UL sub-frames divided between UL timing alignment requests and uplink bandwidth request or
RACH is (RTA - RA) and if RTA - RA is divided on a sub-frame basis the number of UEs in dormant mode at any time is represented by (NTA)-
6. A method as claimed in claim 1 wherein the maximum UEs in dormant mode is watermarked and this water marked value can be changed depending on time averaged peak load (NTA-MAX)-
7. A method as claimed in claim 1 wherein the Number of UEs requiring UL random access is given by (NRA) and the maximum Periodicity of UL Timing alignment procedure is represented as (TTA-YAX) and the minimum periodicity of UL timing alignment procedure is given by (TTA-MIN) where the number of preamble signatures is given by Npreambie.
8. A method as claimed in claim 1 wherein, out of the total number of sub-frames available for UL random access (NUL-RA), a percentage of resources on a sub- frame basis are reserved for UL timing alignment procedure (NTA) depending on the ratio of number of users in dormant mode to number of users in active mode where the percentage is varied dynamically depending on the instantaneous traffic load/random accesses and number of UEs in dormant mode.
9. A method as claimed in claim 1 wherein UEs in uplink dormant mode is allocated a specific sub-frame and a preamble signature for requesting timing alignment. UEs send their respective signatures with/without downlink CQI (Channel Quality Indicator) in case frequency hopping is not used on the DL, or any other relevant information on their assigned uplink resources.
10. A method as claimed in claim 9 wherein in response, base stations send updated timing advance values on the downlink and UEs request uplink bandwidth through the contention based channel, by randomly selecting one of the available sub-frames not reserved for UL timing alignment procedure and randomly select a preamble sequence for transmission.
11. A method as claimed in claim 9 wherein once UE is moved to idle mode or a dedicated UL resource is assigned, the scheduled UL timing alignment sub-frame is re-assigned to another UE who might be entering the UL dormant mode.
12. A method as claimed in claim 1 wherein the allocation of slots for purpose of timing alignment shall be done in a uniform basis each timing alignment slot allocated in a periodicity of TTA-MIN and after the first slots of each TTA-MIN grid is filled, base station starts allocation from the second slot as the number of users in dormant mode increase where the ratio clearly identify the number and location of slots allocated for Timing alignment and random access.
13. A method as claimed in claim 12 wherein by j broadcasting RTA- RA on the broadcast channel to UEs, the entire map can be communicated and the UE on receipt of this ratio can calculate the number and location of RACH slots available to the UE to transmit its random access burst.
14. A method as claimed in claim 1 wherein the distribution of total UL access slots among the timing alignment procedure and random access can be done in fixed patterns known prior to UE and base station where the base station needs to communicate the pattern id currently being used for RACH slot identification.
15. A method as claimed in claim 1 wherein the water filling is done only for certain values of RTA - RA the UE can store only these values and save on the memory required to store all possible patterns .
16. A system for load adaptive timing alignment and uplink random access wherein the said system involves in allocating the frequency-time resources between uplink timing alignment and uplink random access, dynamically based on the ratio of number of user equipments in uplink dormant mode to the number of user equipments intending to perform random access.
17. A method for load adaptive timing alignment and uplink random access substantially described particularly with reference to the accompanying drawings.
18. A system for load adaptive timing alignment and uplink random access substantially described particularly with reference to the accompanying drawings.

Documents:

1523-CHE-2006 AMENDED CLAIMS. 02-07-2013.pdf

1523-CHE-2006 AMENDED PAGES OF SPECIFICATION. 02-07-2013.pdf

1523-CHE-2006 EXAMINATION REPORT REPLY RECEIVED. 02-07-2013.pdf

1523-CHE-2006 FORM-1. 02-07-2013.pdf

1523-CHE-2006 FORM-13. 02-07-2013.pdf

1523-CHE-2006 FORM-5. 02-07-2013.pdf

1523-CHE-2006 OTHER PATENT DOCUMENT. 02-07-2013.pdf

1523-CHE-2006 POWER OF ATTORNEY. 02-07-2013.pdf

1523-CHE-2006 ABSTRACT.pdf

1523-CHE-2006 CLAIMS.pdf

1523-CHE-2006 CORRESPONDENCE OTHERS.pdf

1523-CHE-2006 DESCRIPTION (COMPLETE).pdf

1523-CHE-2006 FORM-1.pdf

1523-CHE-2006 FORM-18.pdf

1523-CHE-2006 FORM-5.pdf

1523-che-2006-correspondnece-others.pdf

1523-che-2006-description(provisional).pdf

1523-che-2006-drawings.pdf

1523-che-2006-form 1.pdf

1523-che-2006-form 26.pdf


Patent Number 256652
Indian Patent Application Number 1523/CHE/2006
PG Journal Number 29/2013
Publication Date 19-Jul-2013
Grant Date 12-Jul-2013
Date of Filing 24-Aug-2006
Name of Patentee SAMSUNG INDIA SOFTWARE OPERATIONS PRIVATE LIMITED
Applicant Address Bagmane Lakeview, Block 'B', No. 66/1, Bagmane Tech Park, C V Raman Nagar, Byrasandra, Bangalore-560093.
Inventors:
# Inventor's Name Inventor's Address
1 PRADEEP DWARAKANATH Samsung India Software Operations Pvt. Ltd., Bagmane Lakeview, Block 'B', No. 66/1, Bagmane Tech Park, C V Raman Nagar, Byrasandra, Bangalore-560093.
PCT International Classification Number H04M03/00
PCT International Application Number N/A
PCT International Filing date
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 NA