Embodiment
Below with reference to accompanying drawing, and in conjunction with specific embodiments the present invention is described in detail.Should be appreciated that the present invention is not limited to specific embodiment.
Fig. 1 is according to a wireless network schematic diagram of the present invention, and this wireless network 1 comprises base station 2 and a plurality of user's set (for simplicity, only demonstrating two user's sets 3 and user's set 4 among the figure).In wireless network 1, user's set 3 and 4 need send resource request information to base station (Node B) 3, to require to distribute certain bandwidth resources.
At first, we discuss the resource scheduling scheme of the logic channel aspect based on user's set of the present invention with one of them user's set 3, also promptly how at user's set 3 places, distribute limited resources at the Different Logic interchannel.
Suppose that at first a logic channel is corresponding to the business with different Q OS.How many resources a logic channel can take is in order to guarantee professional Qos.For Packet data service, a kind of most important Qos information is time delay information.Because the bursty nature of Packet data service, packet data package will arrive user's set at different time.According to time delay, can allow different time delays in the packet data package that difference arrives constantly.Consider this idea, we redefine three new arguments:
TW: dispatching cycle, occupied therebetween resource remains unchanged.
MinR: be to guarantee professional Qos information (time delay information), the minimum transmission rate that user's set is required; Its unit is Bits/TW.
MaxR: peak transfer rate.MaxR is that user's set can be in all transmission of Information speed in buffer of next one transmission logic channel in dispatching cycle.But only be assigned to all logic channels at MinR, this rate requirement just is considered.
Therefore, we say that above-mentioned three parameters of a logic channel are relevant with this professional Qos information (time delay), and should dynamically be calculated as follows in the TW in different dispatching cycles:
TW: the multiple that is transmission time periods TTI.Node B 2 will decide this parameter according to its report of load condition, surplus resources situation and UE, then this parameter be sent to the UE that is scheduled.This can be at Su Shuai ﹠amp; " time " of using in the time scheduling scheme.If TW=1 this means that the speed of logic channel can be dispatched in each TTI.
About latter two parameter, for general and for simplicity, we suppose that user's set 3 has m ongoing logic channel, each channel belong to a kind of business with different Q os we, suppose P
iIndividual grouping is stored in the buffer, and for each grouping, we have following parameter,
A.t
b, be position (or byte) number of b grouping;
B.TO
b, for, according to the Qos information (time delay) of business, b grouping still can be stored in the time (number of TTI) in the buffer
Here, and b ∈ [1, P
i], i ∈ [1, m].
Like this, we know, in order to guarantee professional time delay Qos, the affairs amount that UE must transmit in next transmission time periods TTI (or dispatching cycle TW) is as follows:
Here, 0≤β
i<1 is a factor, and it is relevant with i priority of logical channels and service priority, and this factor can be determined by emulation.Work as β
i=0 o'clock, this means and when carrying out resource request, do not consider priority of logical channels.
Like this, i the needed minimum transmission rate of logic channel of user's set 3 is:
We can see, except priority of logical channels and service priority, and MinR
i lMade up buffer information (l
b) and professional time delay information (TO
b), these information are the resource requirement of presentation logic channel accurately.Therefore, utilize the present invention, can more effective use Limited resources, and help on the logic channel aspect, to avoid " dying of hunger " phenomenon
And for all the preservation information in the buffer that is transmitted in logic channel in next dispatching cycle (TW), i the needed peak transfer rate of logic channel is:
At formula (1)-(3), l
b, b ∈ [1, P
i] be the size of grouping, it is that user's set 3 can obtain.Parameter TO
b, b ∈ [1, P
i] information-related with professional time delay with the due in of grouping.For concrete Packet data service, time delay information can obtain.And the arrival of a grouping is labeled, so TO
b, b ∈ [1, P
i] also can obtain for user's set 3.Has only factor-beta
iCan decide by emulation.In order to simplify, we select β
i=0, because Logic Priority level and service priority will be considered in scheduling process.Therefore, formula (2) is feasible, and is easy to calculate for each logic channel at user's set 3 places.
User's set 3 is obtaining from the resource authorization (SG) of Node B afterwards, and user's set 3 can be according to coming Resources allocation in the Different Logic channel in following manner:
-the order of successively decreasing with priority is each logical channel assignment resource, makes its transmission rate reach MinR as far as possible,
-after the transmission rate of all logic channels all reaches MinR,, then entering for second step if also have resources left, the order of successively decreasing with priority is the further Resources allocation of each logic channel, makes its transmission rate reach MaxR as far as possible,
-last step is if also have resources left, then the logic channel service (as possible property business-best-effort service) of order for still having remaining data of still successively decreasing with priority.
Can be used for the chance in next dispatching cycle, transmitted so that each logic channel has like this.Thereby on the professional Qos of assurance basis, avoid " dying of hunger " phenomenon on the logic channel aspect.
In addition, it is pointed out that above-mentioned formula (2) and (3) represented be transmission rate, but adopt the minimum transmitted in the TW in next dispatching cycle and maximum amount of data, because both are actually and are equal to equally also applicable to the present invention.
Then, we discuss the resource scheduling scheme of the service layer based on user's set of the present invention, also promptly how in user's set 3 places, Resources allocation between different business, to effectively utilize limited resources, guaranteeing on service layer, to avoid " dying of hunger " phenomenon on the professional QOS characteristic basis.
For the scheduling of resource mechanism of above-mentioned logic-based channel aspect, we suppose that each logic channel is mapped to different business.But we know that a plurality of logic channels with equal priority may belong to same business.Service layer rather than logic channel aspect define and Qos information is based on.Therefore, we can come scheduled transmission based on service layer.We know that the logic channel that belongs to identical services may often be mapped to a MAC-d (stream of medium access control-d), and the Different Logic channel that belongs to different business can not be mapped to identical MAC-d stream.As can be seen, be identical with scheduling of resource on the MAC-d aspect in fact for avoiding scheduling of resource that " dying of hunger " phenomenon carries out on the service layer for avoiding " dying of hunger " phenomenon to carry out.
With discussed above the same, supposing for each user's set 3 has n different business to carry out, and we suppose that each MAC-d stream is mapped to a business.For j professional (MAC-d stream), we have following parameter:
-lc
j, belong to the number of the logic channel of j professional (MAC-d stream).For i the logic channel of j professional (MAC-d stream), we have P at supposition
iIndividual grouping is buffered in the buffer, here i ∈ [1, lc
j], and for each grouping, we have l
bBe the bit number of b grouping,
TO
b, being b grouping still can be buffered in time in the buffer according to the QOS parameter (time delay information) of business.Here, and b ∈ [1, P
i].
Like this, we know that in order to guarantee professional time delay QOS, j business needs following speed in the TW in next dispatching cycle:
Here, 0≤β
j<1 is a factor relevant with service priority, and can determine by emulation.As mentioned above, β
j=0 means when request resource requires, and does not consider service priority.And service priority information will be considered in scheduling step subsequently.Therefore, we know MinR
j sBe that j business is in next TW (or TTI), in order to ensure its time delay Qos, needed minimum-rate.And speed is for all groupings in the buffer that is transmitted in j business, also promptly:
Like this, after obtaining resource authorization by Node B 2, user's set 3 comes Resources allocation in different business (MAC-d stream) according to following manner:
-be each business (MAC-d stream) Resources allocation in proper order with successively decreasing of service priority, be that the transmission rate of each business (MAC-d stream) reaches MinR as far as possible
j s,
-all reached MinR when the transmission rate of all business (MAC-d stream)
j sIf resource also has residue, then entered for second step, be further Resources allocation of each business (MAC-d stream) in proper order with successively decreasing of service priority, make its transmission rate reach MaxR as far as possible
j s,
-last step is if still have resources left, and then the order of still successively decreasing with priority is business (MAC-d stream) service that still the has remaining data (property of doing the best business-best-effortservice).
We we can say that each business (MAC-d stream) has and be used for the chance transmitted in next dispatching cycles.Thereby avoid " dying of hunger " phenomenon on service layer, and professional QOS delay character is guaranteed also
Fig. 2 is according to of the present invention, and user's set 3 carries out the method flow diagram of resource allocation in the logic channel aspect.
Here, the topological structure schematic diagram in conjunction with wireless network shown in Figure 1 comes whole scheduling process is described.Suppose that user's set 3 has m logic channel that is carrying out (on going), and each logical mappings is to a kind of business with different delay.For i logic channel, suppose P
iIndividual grouping is stored in the buffer, and each grouping has following parameter:
A) l
b, be position (or byte) number of b grouping;
B) TO
b, for, according to the Qos information (time delay) of business, b grouping still can be stored in the time (number of TTI) in the buffer
Here, and b ∈ [1, P
i], i ∈ [1, m].
In step 201, for each logic channel i, user's set 3 obtains parameters such as the time that reaches, time delay information of corresponding grouping number Pi, each grouping in buffer memory.
In step 202, user's set is according to the time of advent of each grouping in the buffer memory and the time delay information of corresponding service, and user's set calculates this grouping still can be stored in time T O in the buffer memory
b
In step 203, user's set calculates by following formula (2), in order to guarantee professional time delay Qos, i logic channel required minimum transmission rate in next one TW dispatching cycle:
Here, 0≤β
i<1 is a factor, and it is relevant with i priority of logical channels and service priority, and this factor can be determined by emulation.Work as β
i=0 o'clock, mean and when carrying out resource request, do not consider the Logic Priority level.
In step 204, user's set calculates by following formula (3), and for all the preservation information in the buffer that is transmitted in logic channel at the next one dispatching cycle among the TW, i the needed maximum rate of logic channel is:
In step 205, user's set 3 is with resource request signaling SI notice Node B (base station) 2.
In step 206, after the resource authorization SG that obtains from Node B, user's set 3 is each logical channel assignment resource according to the order of successively decreasing of priority of logical channels, makes its transmission rate reach MinR as far as possible
In step 207, user's set judges whether whether the transmission rate of each logic channel reaches MinR and still have unnecessary resource to be allocated, if then enter next step 208; If not, ending resource assigning process then.
In step 208, user's set 3 is the further Resources allocation of each logic channel according to the order of successively decreasing with priority, makes its transmission rate reach MaxR as far as possible.
In step 209, user's set 3 judges whether to still have logic channel to have remaining data, and still has resources left, if then enter next step 210; If not, ending resource assigning process then.
In step 210, the logic channel service (as possible property business) of order that user's set 3 still successively decreases with priority for still having remaining data.
Fig. 3 is according to of the present invention, and user's set 3 carries out the method flow diagram of scheduling of resource in service layer.
At this, still the topological structure schematic diagram in conjunction with wireless network shown in Figure 1 comes whole scheduling of resource process is described.Suppose that user's set 3 has m logic channel that is carrying out (on going), and each logical mappings is to a kind of business with different delay.For i logic channel, suppose P
iIndividual grouping is stored in the buffer, and each grouping has following parameter:
A) l
b, be position (or byte) number of b grouping;
B) TO
b, for, according to the Qos information (time delay) of business, b grouping still can be stored in the time (number of TTI) in the buffer
Here, and b ∈ [1, P
i], i ∈ [1, m].
In step 301, for each logic channel i, user's set 3 obtains parameters such as the time that reaches, time delay information of corresponding grouping number Pi, each grouping in buffer memory.
In step 302, user's set is according to the time of advent of each grouping in the buffer memory and the time delay information of corresponding service, and user's set calculates this grouping still can be stored in time T O in the buffer memory
b
In step 303, user's set calculates by following formula (4), and in order to guarantee professional time delay Qos, j professional (MAC-d stream) is required minimum transmission rate in next one TW dispatching cycle:
Here, 0≤β
j<1 is a factor relevant with service priority, and can determine by emulation.As mentioned above, β
j=0 means when request resource requires, and does not consider service priority.And service priority information will be considered in scheduling step subsequently.
In step 304, user's set calculates by following formula (5), and in order to be transmitted in j all preservation information in the professional buffer among the TW in next dispatching cycle, j professional (MAC-d stream) needed maximum rate is:
In step 305, user's set 3 is with resource request signaling SI notice Node B (base station) 2.
In step 306, after the resource authorization SG that obtains from Node B, user's set 3 is each traffic assignments resource according to the order of successively decreasing of professional (MAC-d stream) priority, makes its transmission rate reach MinR as far as possible
In step 307, user's set judges whether whether each professional transmission rate reaches MinR and still have unnecessary resource to be allocated, if then enter next step 308; If not; Ending resource assigning process then.
In step 308, user's set 3 for each professional further Resources allocation, makes its transmission rate reach MaxR according to the order of successively decreasing with priority as far as possible.
In step 309, user's set 3 judges whether whether each professional transmission rate reaches MaxR and still have unnecessary resource to be allocated, if then enter next step 310; If not, ending resource assigning process then.
In step 310, if still have resources left, business (MAC-d stream) service (as possible property business) of order that user's set 3 still successively decreases with priority for still having remaining data.
It should be noted that, what describe in conjunction with Fig. 2 and 3 above is a preferred embodiment of the present invention, to be user's sets combine the Qos information of business and logic channel or professional buffer information in logic channel aspect or service layer main points of the present invention, comes the request of expressing bandwidth resources to Node B (base station) 2.To the consideration such as other information of priority of logical channels, service priority etc. all is to belong to optional feature.
Fig. 4 is the block diagram according to user's set of the present invention.Below, in conjunction with Fig. 1 user's set is specifically described.This user's set 3 is arranged in a wireless network 1, links to each other with wireless network by NodeB (base station) 2, and when needs transmission information, need be to the certain bandwidth resources of Node B (base station) 2 requests.
This user's set 3 comprises a buffer 41, calculation element 42, a controller 43 and a storage device 44.In buffer 41, store the data that will send, and this user have a plurality of ongoing logic channels, the corresponding a kind of business of each logic channel.Storage is by each logic channel and professional precedence information in storage device 44.
At first, we discuss the preferred embodiment that carries out the user's set 3 of scheduling of resource in the logic channel aspect.
Particularly, suppose that user's set 3 has m logic channel that is carrying out (on going), and each logical mappings is to a kind of business with different delay.For i logic channel, suppose P
iIndividual grouping is stored in the buffer, and each grouping has following parameter:
A) l
b, be position (or byte) number of b grouping;
B) TO
b, for, according to the Qos information (time delay) of business, b grouping still can be stored in the time (number of TTI) in the buffer
Here, and b ∈ [1, P
i], i ∈ [1, m].
When needs carried out scheduling of resource, for i logic channel, control device 43 need be by obtaining corresponding grouping number Pi, the time that reaches of each grouping, each professional time delay information in buffer memory.And control device 43 is also by the precedence information that obtains each professional relevant information (being preferably time delay Qos) and each logic channel in the storage device 44.
Control device 43 is given calculation element 42 with these message transmission, and is calculated by following formula (2) by calculation element 42, in order to guarantee professional time delay Qos, i logic channel required minimum transmission rate in next one TW dispatching cycle:
Here, 0≤β
i<1 is a factor, and it is relevant with i priority of logical channels and service priority, and this factor can be determined by emulation.Work as β
i=0 o'clock, mean and when carrying out resource request, do not consider the Logic Priority level.
Then, calculation element also calculates by following formula (3), and for all the preservation information in the buffer that is transmitted in logic channel at the next one dispatching cycle among the TW, i the needed maximum rate of logic channel is:
Calculation element 42 is transferred to control device 43 with result of calculation.Then, control device 43 control user's sets 3 are with resource request signaling SI notice Node B (base station) 2.
After the resource authorization SG that obtains from Node B, control device 3 can also basis come Resources allocation in the Different Logic channel in following manner:
-be in proper order each logical channel assignment resource to make its transmission rate reach MinR with successively decreasing of priority of logical channels as far as possible,
-after the transmission rate of all logic channels all reaches MinR,, then entering for second step if also have resources left, the order of successively decreasing with priority is the further Resources allocation of each logic channel, makes its transmission rate reach MaxR as far as possible,
-last step is if also have resources left, then the logic channel service (as possible property business) of order for still having remaining data of still successively decreasing with priority.
Secondly, we discuss the preferred embodiment that carries out the user's set 3 of scheduling of resource in service layer.
When needs transmission information, control device 43 is at first by the current state information that obtains buffer in the buffer 41 (also promptly wherein institute's stored packet data volume), and by obtaining each professional relevant information (being preferably time delay Qos) and each professional precedence information in the storage device 44.
Particularly, supposing for each user's set 3 has n different business to carry out, and for j professional (MAC-d stream), we have following parameter:
-lc
j, belong to the number of the logic channel of j professional (MAC-d stream).For i the logic channel of j professional (MAC-d stream), we have P at supposition
iIndividual grouping is buffered in the buffer, here i ∈ [1, lc
j], and for each grouping, we have
-l
bBe the bit number of b grouping,
-TO
b, being b grouping still can be buffered in time in the buffer according to the QOS parameter (time delay information) of business.Here, and b ∈ [1, P
i].
When needs carried out scheduling of resource, for j professional (MAC-d stream), control device 43 need be by obtaining corresponding grouping number Pi, the time that reaches of each grouping, each professional parameter such as time delay information in buffer memory in the buffer.And control device 43 is also by the precedence information that obtains each business (MAC-d stream) in the storage device 44.
Control device 43 is given calculation element 42 with these message transmission, and is calculated by following formula (4) by calculation element 42, in order to guarantee professional time delay Qos, j business required minimum transmission rate in next one TW dispatching cycle:
Here, 0≤β
j<1 is a factor relevant with service priority, and can determine by emulation.As mentioned above, β
j=0 means when request resource requires, and does not consider service priority.And service priority information will be considered in scheduling step subsequently.Therefore, we know MinR
j sBe that j business is in next TW (or TTI), in order to ensure its time delay Qos, needed minimum-rate.And speed is for all groupings in the buffer that is transmitted in j business, also promptly:
Calculation element 42 is transferred to control device 43 with result of calculation.Then, control device 43 control user's sets 3 are with resource request signaling SI notice Node B (base station) 2.
After the resource authorization SG that obtains from Node B, control device 3 can also basis come Resources allocation in different business in following manner:
-be each business (MAC-d stream) Resources allocation in proper order with successively decreasing of service priority, be that the transmission rate of each business (MAC-d stream) reaches MinR as far as possible
j s,
-all reached MinR when the transmission rate of all business (MAC-d stream)
j sIf, also have resources left, then entered for second step, be further Resources allocation of each business (MAC-d stream) in proper order with successively decreasing of service priority, make its transmission rate reach MaxR as far as possible
j s,
-last step is if also have resources left, and then the order of still successively decreasing with priority is business (MAC-d stream) service that still the has remaining data (property of doing the best business-best-effortservice).
More than specific embodiments of the invention are described.Need to understand being, the present invention is not limited to above-mentioned specific for execution mode, and those skilled in the art can make various distortion or modification within the scope of the appended claims.