Patent Application
20090125778
The present invention relates to a communication system that uses an Automatic Repeat reQuest (ARQ) method in which the reception side automatically makes a request to the transmission side that transmission data should be re-transmitted, and also relates to a transmission-side communication apparatus and a reception-side communication apparatus that are included in such a communication system.
Various types of the ARQ method mentioned above have conventionally been considered, and in particular, typical examples are as follows:
(1) Stop-and-wait (SAW) ARQ method
(2) Go-Back-N (GBN) ARQ method
(3) Selective Repeat (SR) ARQ method
The SAW_ARQ method is characterized in that a delivery confirmation is made for each transmission block, and a new block is not transmitted until an ACK is returned from the reception side. The SAW_ARQ method is also used in a Media Access Control (MAC) layer according to the Institute of Electrical and Electronics Engineers, Inc. (IEEE) 802.11 standard. Although the SAW_ARQ method is simple, it is disadvantageous in that the transmission efficiency is low, and user throughput relative to the capacity of the communication line is not sufficient.
The GBN_ARQ method is characterized in that transmission blocks continue to be transmitted even if no ACK is received from the reception side, but when an NACK is returned from the reception side, the continual transmission is resumed from a corresponding sequence number. Although the GBN_ARQ method is also simple, it is disadvantageous in that the transmission efficiency is significantly degraded in a communication environment like a wireless communication line where communication errors occur frequently.
In contrast, according to the SR_ARQ method, only the blocks in which an error has been detected on the reception side are re-transmitted. The SR_ARQ method is used as the ARQ method according to IEEE 802.16. The SR_ARQ method has advantageous characteristics where the transmission efficiency is high, and also, compared to the SAW_ARQ method and the GBN_ARQ method, it has a capability of preventing the user throughput from being degraded drastically, because the reception window is updated whenever it is necessary in correspondence with each piece of ACK/NACK information.
According to the basic ARQ methods explained above, when a reception error has occurred, the same data is re-transmitted as a means for recovering from the error. Thus, when the error rate of the transmission path becomes worse, the number of times the re-transmission process is performed increases. Thus, the throughput is significantly degraded. To improve the throughput for a transmission path having a bad error rate, an error correction code or an erasure correction code is used together with an ARQ method.
For instance, a representative example of an erasure correction code according to a conventional technique is a Luby Transfer (LT) code. A communication method that uses the LT code has a number of advantageous characteristics where a tentative communication path called an erasure communication path is set up, and a packet having a code length of n can arbitrarily be encoded on the transmission side with respect to a packet having an information length of k while n>k is satisfied, whereas as many information packets as k can successfully be decoded on the reception side even if only as few packets as n+ε at most (where ε≈1.05×n to 1.2×n) have successfully been received (see, for example, Non-patent Document 1).
Even if the SR_ARQ method that has the highest throughput performance among the basic ARQ methods explained above is used as a means for controlling the re-transmission for a wireless communication line having a high speed and a large capacity, in a case where, for example, the communication line is in such a state that a re-transmission request is frequently made (including situations affected by delay in monitoring and adaptive control of communication line state information), or in a case where the system includes an uplink communication line used for a delivery confirmation that has a low level of performance although a downlink communication line having a large capacity is available, a problem arises, for example, the user throughput is significantly degraded temporarily because the update of the ARQ transmission window is delayed.
In view of the problems explained above, it is an object of the present invention to provide a communication system, a transmission-side communication apparatus, and a reception-side communication apparatus that are able to avoid or inhibit a temporary and drastic degradation of the user throughput, even when they are applied to, for example, a communication system that has a possibility of experiencing such a state of communication line in which a re-transmission request is frequently made.
To solve the problems as described above and to achieve an object, a communication system according to the present invention is a communication system in which a reception-side communication apparatus makes a request to a transmission-side communication apparatus that a transmission data signal be re-transmitted, wherein the transmission-side communication apparatus includes a transmission scheduling unit that determines a transmission amount to be transmitted, at least, to the reception-side communication apparatus, an erasure correction encoding unit that performs an erasure correction encoding process on an information packet group that is made up of a plurality of packets to be transmitted so as to generate one or more erasure correction coded packets that fit the transmission amount instructed by the transmission scheduling unit and specifies the one or more erasure correction coded packets as a unit of delivery confirmation, and a transmitting unit that transmits the transmission data signal that has been generated by performing a predetermined modulation process on each of the erasure correction coded packets, and the reception-side communication apparatus includes an erasure correction decoding unit that generates the information packet group by performing an erasure correction decoding process on the transmission data signal that has been received and generates, in a case where the erasure correction decoding process has successfully been performed on the transmission data signal, a reception completion signal indicating that reception of the transmission signal has been completed for each unit of delivery confirmation, and a transmitting unit that transmits a delivery confirmation signal that has been generated based on the reception completion signal.
According to the present invention, in the transmission-side communication apparatus, the erasure correction encoding process is performed on the information packet group that is made up of the plurality of packets to be transmitted, so that the one or more erasure correction coded packets that fit the predetermined transmission amount are generated. The one or more erasure correction coded packets are specified as one of units in which a delivery confirmation is made (hereinafter “a unit of delivery confirmation”) and are transmitted to the reception-side communication apparatus. In the reception-side communication apparatus, the information packet group is generated by performing the erasure correction decoding process on the received signal. In the case where the erasure correction decoding process has successfully been performed, the delivery confirmation signal indicating that reception of the transmission data signal has been completed is generated for each unit of delivery confirmation and transmitted to the transmission-side communication apparatus. Thus, even if the present invention is applied to a communication system that has a possibility of experiencing such a state of communication line in which a re-transmission request is frequently made, an advantageous effect is achieved where it is possible to avoid or inhibit a temporary and drastic degradation of the user throughput.
Exemplary embodiments to illustrate a communication system, a transmission-side communication apparatus, and a reception-side communication apparatus according to the present invention will be explained in detail with reference to the accompanying drawings. The present invention is not limited to these exemplary embodiments.
In
In the transmission-side communication apparatus 11, IP packets 31 are input to the data storing unit 111, and the IP packets 31 are output from the modulating unit 115. Also, a delivery confirmation signal (ACK) 42 that has been transmitted via a communication path 51 is input to the demodulating unit 116.
In the reception-side communication apparatus 21, a transmission data signal 41 that has been transmitted via the communication path 51 is input to the demodulating unit 211, and IP packets 32 are output from the IP packet reproducing unit 214 to, for example, an application or another communication apparatus (not shown). Also, the delivery confirmation signal (ACK) 42 is output from the modulating unit 216 of the reception-side communication apparatus 21.
In the explanation below, the example of the configuration as shown in
In the configuration shown in
Also, in the configuration shown in
Next, an operation of the communication system according to the first embodiment will be explained with reference to
“The predetermined level of data amount” mentioned above denotes, for example, a data amount expressed with a value obtained by subtracting the data amount of header information (e.g., padding size information) from the value of K×Lmax, where K is the number of information packets before an erasure correction encoding process is performed, and Lmax is the maximum length of the information packet.
In the process described above, in the case where the process to store the IP packets into the data storing unit 111 is performed based on “the predetermined level of data amount”, the data storing unit 111 forwards, for example, the data that has been stored therein and the header information to the erasure correction encoding unit 112 after dividing the data and the header information into as many information packets as K, while each of the information packets has a length of L=Lmax.
In the case where the IP packets that have been stored in the data storing unit 111 has not reached “the predetermined level of data amount”, the process to store the IP packets is performed until “the predetermined period of time has elapsed”. In this situation, the data storing unit 111 forwards the stored data to the erasure correction encoding unit 112 after dividing the data into as many information packets as K, while each of the information packets has a length of L and is padded so as to have a size of K×L, where the length L is determined so that the sum P of the data amount of the stored data and the amount of the header information satisfies P<K×L.
The erasure correction encoding unit 112 stores the group of information packets (hereinafter, “information packet group”) that has been received from the data storing unit 111 into a buffer 112a included therein. The transmission scheduling unit 113 determines a modulation method for each user based on the information such as a Carrior to Noise Ratio (CRN) and a Bit Error Rate (BER) for the connection. The transmission scheduling unit 113 also calculates a transmission amount for each connection. In addition, the erasure correction encoding unit 112 generates, based on an erasure correction code, a number of coded packets that fits within the range of the transmission amount determined and instructed by the transmission scheduling unit 113 and forwards the generated coded packets to the error correction encoding unit 114. In this situation, a packet header containing sequence numbers showing the order of packet generation and the length L of the packets is appended to the generated erasure correction coded packets. Further, a Cyclic Redundancy Check (CRC) code that is used on the reception side to judge whether the packet has successfully been received or not is also appended to the generated erasure correction coded packets. The size of the packet header can be can be reduced if, inn this situation, for example, a piece of one-bit information is used for judging whether the packet length L is Lmax or not, and the length L is not appended when L=Lmax is satisfied. Also, it is preferable to have an arrangement in which the sequence numbers are reset for each information packet group forwarded from the data storing unit 111.
The error correction encoding unit 114 forwards the erasure correction coded packets that have been received from the erasure correction encoding unit 112 to the modulating unit 115, after performing an error correction encoding process thereon. The modulating unit 115 performs a digital modulation process according to a modulation method such as Binary Phase Shift Keying (BPSK), Quadrature Phase Shift Keying (QPSK), or multi-value Quadrature Amplitude Modulation (QAM). The modulating unit 115 then transmits a generated modulation signal as the transmission data signal 41 to the reception-side communication apparatus 21 via the communication path 51.
The transmission data signal 41 transmitted from the transmission-side communication apparatus 11 is input to the reception-side communication apparatus 21. The demodulating unit 211 performs a digital demodulation process on the transmission data signal 41, based on the modulation method that has been applied to the transmission data signal 41 (e.g., a BPSK, a QPSK, or a multi-value QAM), and forwards the demodulated data to the error correction decoding unit 212. The error correction decoding unit 212 receives the demodulated data from the demodulating unit 211 and performs an error correction decoding process thereon. The error correction decoding unit 212 then forwards the result to the erasure correction decoding unit 213.
The erasure correction decoding unit 213 judges whether the received packets have properly been received, based on the CRC code. In the case where the received packets have properly been received, the erasure correction decoding unit 213 stores the received packets that have been input thereto, into a buffer 213a in the order indicated by the sequence numbers. When the number of received packets becomes equal to or larger than the number of information packets specified on the transmission side (specified as K in the first embodiment), the erasure correction decoding unit 213 performs an erasure correction decoding process by using all of the received packets stored in the buffer 213a. Further, in the case where the erasure correction decoding process has successfully been performed, the erasure correction decoding unit 213 clears all of the received packets from the buffer 213a and sequentially forwards the information packet group that has been decoded, to the IP packet reproducing unit 214. Further, the erasure correction decoding unit 213 generates a reception completion signal (ACK). The error correction encoding unit 215 performs an error correction encoding process thereon. Subsequently, the modulating unit 216 performs a digital modulation process based on the predetermined modulation method, so that the modulation signal is transmitted as a delivery confirmation signal (ACK) 42 to the transmission side via the communication path 51. On the other hand, in the case where the erasure correction decoding process has failed, the erasure correction decoding unit 213 continues to receive the received packets until the decoding process is successfully performed.
The IP packet reproducing unit 214 connects together the information packet group that has been received from the erasure correction decoding unit 213 and extracts IP packets by referring to the information such as the header information (e.g., the padding size information) and the length information of the IP packets. The IP packet reproducing unit 214 then performs a process of, for example, forwarding the extracted packets to an application layer or transferring the extracted packets to another communication apparatus, as the IP packets 32. In the case where the information that is required in the generation of the IP packets is partially contained in both the present information packet group and the next information packet group, the data that is received first is stored so that, after the next information packet group has been received, the packets are connected together so as to reproduce the desired IP packets.
The delivery confirmation signal 42 that has been transmitted from the reception-side communication apparatus 21 is returned to the transmission-side communication apparatus 11. The demodulating unit 116 forwards the returned delivery confirmation signal 42 to the error correction decoding unit 117, after performing a digital demodulation process thereon. Having received the demodulated data from the demodulating unit 116, the error correction decoding unit 117 performs an error correction decoding process thereon and forwards the result to the erasure correction encoding unit 112. Having received the delivery confirmation signal, the erasure correction encoding unit 112 clears the buffer of the information packet group on which an encoding process is currently being performed and sets the next information packet group that has been received from the data storing unit 111 as the target of an encoding process. Thus, until the transmission-side communication apparatus 11 receives the delivery confirmation signal, the transmission-side communication apparatus 11 assumes that the information that is currently being transmitted has not been completely received by the reception-side communication apparatus 21 and continues to generate and transmit redundancy packets (i.e., erasure correction coded packets) to the reception-side communication apparatus 21.
As explained above, the transmission-side communication apparatus 11 generates new erasure correction coded packets while incrementing the sequence numbers, until the transmission-side communication apparatus 11 receives the delivery confirmation signal. However, it is not preferable to continue to generate the erasure correction coded packets in a limitless manner. Thus, for example, an arrangement is desirable in which a threshold value (i.e., an upper limit value) is specified based on a temporal aspect or a quantitative aspect, so that when the judgment element has reached the threshold value, the delivery confirmation process is discontinued and the transmission buffer is cleared, and also, the transmission is resumed with a next information packet group.
Also, it is a good idea to have an arrangement in which, in the case where the reception-side communication apparatus 21 has received an erasure correction coded packet from the transmission communication apparatus 11 after having transmitted a delivery confirmation signal 42 to the transmission-side communication apparatus 11, and if, for example, a large sequence number is being used, the reception-side communication apparatus 21 may discard the received packets and periodically notify the transmission-side communication apparatus 11 with a delivery confirmation signal. It is also a good idea to have another arrangement in which, when a predetermine period of time has elapsed after the delivery confirmation signal 42 is transmitted, the reception-side communication apparatus 21 may discard the received erasure correction coded packets and notify the transmission-side communication apparatus 11 with a delivery confirmation signal every time the erasure correction coded packets have been received. After that, it is also a good idea to have an arrangement in which, when erasure correction coded packets of which the sequence numbers have gone back to smaller numerical values have been received, the reception-side communication apparatus 21, recognizing that the transmission of the next information packet group has been started, starts storing the received packets into the buffer and stops notifying the transmission-side communication apparatus 11 with the delivery confirmation signal.
As explained above, in the communication system according to the first embodiment, it is understood on the transmission side that when no delivery confirmation signal is received, it implicitly means that re-transmission is requested so that the redundancy packets continue to be transmitted to the reception side. With this arrangement, even if a delivery confirmation is made for each information packet group, it is possible to reduce, by a large amount, the feedback information to be transmitted to the transmission side without degrading the user throughput significantly.
Also, in the communication system according to the first embodiment, it is not necessary to have the functions of a window controlling unit and a state management unit that are required when, for example, the SR_ARQ method is used. Thus, it is possible to simplify the control and make the scale of the circuit smaller.
In addition, because the communication system according to the first embodiment has no restriction related to full duplex or half duplex, it is possible to apply the communication system to any communication method.
It is possible to modify a part of the first embodiment as described below, from the following aspects:
(1) the size of the packet headers;
(2) the delay amount related to the storing of the data;
(3) compatibility with a plurality of modulation methods (i.e., adaptive modulation control); and
(4) the throughput related to the delivery confirmation.
As shown in
In the example shown in
Let us discuss a situation in which the erasure correction encoding unit 112 included in the transmission-side communication apparatus 11 forwards erasure correction coded packets to the error correction encoding unit 114. In this situation, the erasure correction encoding unit 112 packs the erasure correction coded packets according to the physical layer capacity 83. On the other hand, since the packet length of each of the erasure correction coded packets that constitute the frame 80 is predetermined, it is possible to reduce the header size by, when packing the packets, putting only the sequence number of the erasure correction coded packet positioned at the head into the packet header, as shown in
When the packing technique as described above is used, it is possible to increase the effect of reducing the header size because only one CRC code needs to be appended to the packing data. Additionally, when the packing technique described above is used, by putting the information of the packet length of the erasure correction coded packets into, for example, the header portion 81, it is possible to deliver this information to the destination of the communication. Also, as shown in the lower half of
As explained above, in consideration of occurrence of transmission delay related to the storing of the data, it is possible to adjust the delay amount related to the storing of the data by increasing or decreasing, according to the QoS class, the number of erasure correction coded packets corresponding to the coding rate of “1”, without having to change the packet size of each of the erasure correction coded packets. Also, it is possible to adjust the delay amount related to the storing of the data by increasing or decreasing, according to the QoS class, the packet size of each of the erasure correction coded packets, without having to change the number of erasure correction coded packets corresponding to the coding rate of “1”.
According to the first embodiment, as explained above, the instructions related to the modulation method for each user and the transmission amount for each connection are output from the transmission scheduling unit 113 to the erasure correction encoding unit 112. In this situation, it is possible to easily perform the scheduling in such a manner that, for example, as many erasure correction coded packets as X at the beginning are assigned to the modulation method called 64 QAM, whereas as many erasure correction coded packets as Y that follow are assigned to the modulation method called 16 QAM, and erasure correction coded packets that further follow are assigned to the modulation method called QPSK. In other words, it is possible to easily perform the scheduling so as to lower the degree of modulation and raise the degree of redundancy on the assumption that the larger the number of packets being transmitted is, the lower the quality of communication becomes. Thus, it is possible to simplify the functions of the adaptive modulation control. Also, in some situations, it is possible to omit the functions of the adaptive modulation control themselves.
As described above, according to the present embodiment, in the transmission-side communication apparatus, the erasure correction encoding process is performed on the information packet group that is made up of the plurality of packets to be transmitted, so that the one or more erasure correction coded packets that fit the predetermined transmission amount are generated. The one or more erasure correction coded packets are specified as one of units in which a delivery confirmation is made (hereinafter “a unit of delivery confirmation”) and are transmitted to the reception-side communication apparatus. In the reception-side communication apparatus, the information packet group is generated by performing the erasure correction decoding process on the received signal. In the case where the erasure correction decoding process has successfully been performed, the delivery confirmation signal indicating that reception of the transmission data signal has been completed is generated for each unit of delivery confirmation and transmitted to the transmission-side communication apparatus. Thus, even if the present invention is applied to a communication system that has a possibility of experiencing such a state of communication line in which a re-transmission request is frequently made, an advantageous effect is achieved where it is possible to avoid or inhibit a temporary and drastic degradation of the user throughput.
Also, when the techniques according to the first embodiment are used, the re-transmission process performed by the transmission-side communication apparatus is equivalently substituted by an additional transmission of the erasure correction coded packets. As a result, the frequency with which the delivery confirmation signals are transmitted to the feedback channel and the frequency with which the ARQ transmission window is updated are significantly lowered. Thus, for these reasons also, it is possible to avoid or inhibit temporal and drastic degradation of the user throughput.
In addition, it is possible to avoid degradation of the user throughput more effectively by arranging the size of the unit of delivery confirmation to be large compared to the packet size of each of the erasure correction coded packets, or by arranging the packet size of each of the erasure correction coded packets to be sufficiently small compared to the size of the unit of delivery confirmation.
Similarly, the basic configuration of the reception-side communication apparatus 21 is the same as that according to the first embodiment; however, the constituent elements such as the demodulating unit 211, the error correction decoding unit 212, the error correction encoding unit 215, and the modulating unit 216 are able to perform communication based on a plurality of channels that use mutually the same communication access method, a plurality of communication access methods, or a combination thereof. In some situations, another arrangement is acceptable in which the constituent elements such as the error correction decoding unit 212 and the error correction encoding unit 215 are shared between the plurality of channels or between the plurality of communication access methods.
In
The number of constituent elements such as the demodulating unit 116 and the error correction decoding unit 117 that are included in the transmission-side communication apparatus 11 and that process delivery confirmation signals 42 and 44 as well as the error correction encoding unit 215 and the modulating unit 216 that are included in the reception-side communication apparatus 21 and that generate the delivery confirmation signals 42 and 44 does not necessarily have to be equal to the number of constituent elements such as the error correction encoding unit 114 and the modulating unit 115 that are included in the transmission-side communication apparatus 11. Another arrangement is acceptable in which only one constituent elements or a smaller number of constituent elements are included.
Next, an operation of the communication system according to the second embodiment will be explained, with reference to
First, like in the operation according to the first embodiment, in the transmission-side communication apparatus 11, the erasure correction encoding unit 112 receives an information packet group from the data storing unit 111 and stores the received information packet group into the buffer 112a. Subsequently, the transmission scheduling unit 113 determines a transmittable amount for each channel or each communication access method. The erasure correction encoding unit 112 generates, based on an erasure correction code, a number of packets that fit the transmission amount corresponding to each channel or each communication access method as instructed by the transmission scheduling unit 113. The erasure correction encoding unit 112 then forwards the generated packets to the error correction encoding unit 114. In this situation, as for the sequence numbers assigned to the erasure correction coded packets, it is preferable to have an arrangement in which, for example, the sequence numbers are used in common between the plurality of channels or the plurality of communication access methods, instead of assigning independent sets of sequence numbers respectively, so that mutually different sets of erasure correction coded packets are transmitted for mutually the same information packet group to be encoded even between the plurality of channels or the plurality of communication access methods.
In the reception-side communication apparatus 21, when having properly received the erasure correction coded packets, the erasure correction decoding unit 213 stores the received packets that have been input thereto, into the buffer 213a according to the order indicated by the sequence numbers, without being concerned that the channels are mutually different or that the communication access methods are mutually different and performs an erasure correction decoding process. When having successfully performed the erasure correction decoding process, the erasure correction decoding unit 213 forwards the decoded information packet group to the IP packet reproducing unit 214 and also gives information that is required in the generation of the delivery confirmation signals 42 and 44 to the error correction encoding unit 215. The error correction encoding unit 215 performs an error correction encoding process on the information, before the information is output from the modulating unit 216 through a communication access method or a channel having good communication line quality or through a fixed communication access method or a fixed channel. The processes that are performed thereafter are the same as those according to the first embodiment. Thus, the explanation thereof will not be repeated.
As explained above, in the communication system according to the second embodiment, the erasure correction coded packets are transmitted while being distributed in the plurality of channels or the plurality of communication access methods. After that, when the erasure correction decoding process is performed in the reception-side communication apparatus, the distributed erasure correction coded packets are put together. Thus, it is possible to achieve a diversity effect between the mutually different channels or the mutually different communication access methods. In addition, even if such a method is used, the reception-side communication apparatus is able to process the data without being concerned that the channels are mutually different or that the communication access methods are mutually different. Thus, it is possible to achieve the advantageous effect of preventing or inhibiting an increase in the delay caused by a transmission delay difference between the channels or the communication access methods and preventing or inhibiting the degradation of the user throughput. In addition, it is possible to achieve an advantageous effect where the reception-side communication apparatus does not need to control the order of the received packets.
The configuration shown in
Furthermore, the transmission-side communication apparatus 11 and the transmission-side communication apparatus 12 may use mutually the same communication system or mutually different communication systems. However, in the case where they use mutually different communication systems, it is also necessary to configure the reception-side communication apparatus 21 so as to include the demodulating unit 211, the error correction decoding unit 212, the error correction encoding unit 215, and the modulating unit 216 that are compatible with both of the mutually different communication systems.
Next, an operation of the communication system according to the third embodiment will be explained. The basic flow of the processes is the same as the one according to the first embodiment. Thus, only the part that is different from the first embodiment will be mainly explained.
In
Subsequently, the erasure correction encoding unit 112 included in the transmission-side communication apparatus 11 and the erasure correction encoding unit 112 included in the transmission-side communication apparatus 12 perform the encoding process after changing the starting number of the sequence numbers assigned to the erasure correction coded packets so that the sequence numbers do not overlap between these transmission-side communication apparatuses. The operation that is performed thereafter is the same as the one according to the first embodiment. The transmission data signals 41 and 43 that are generated in the transmission-side communication apparatus 11 and 12, respectively, are transmitted to the reception-side communication apparatus via the communication path 51 and 52, respectively.
In the reception-side communication apparatus 21, when having properly received the erasure correction coded packets, the erasure correction decoding unit 213 stores the received packets that have been input thereto, into the buffer 213a in the order indicated by the sequence numbers, without being concerned that the packets have been transmitted from the mutually different transmission-side communication apparatuses. The erasure correction decoding unit 213 then performs an erasure correction decoding process. When having successfully performed the erasure correction decoding process, the erasure correction decoding unit 213 forwards the decoded information packet group to the IP packet reproducing unit 214 and also gives information that is required in the generation of the delivery confirmation signals 42 and 44 to the error correction encoding unit 215. The error correction encoding unit 215 performs an error correction encoding process on the information, before the information is transmitted to both the transmission-side communication apparatus 11 and the transmission-side communication apparatus 12. The processes that are performed thereafter are the same as those according to the first embodiment. Thus, the explanation thereof will not be repeated.
As explained above, in the communication system according to the third embodiment, the erasure correction coded packets are transmitted while being distributed in the plurality communication apparatuses. After that, when the erasure correction decoding process is performed in the reception-side communication apparatus, the distributed erasure correction coded packets are put together. Thus, it is possible to achieve a diversity effect between the mutually different transmission-side communication apparatuses.
Next, an operation of the communication system according to the fourth embodiment will be explained. The basic flow of the processes is the same as the one according to the first embodiment. Thus, only the part that is different from the first embodiment will be mainly explained.
Let us assume that the mobile communication terminal 93 is currently performing predetermined wireless communication with the handover origin wireless base station 91, based on the functions described above. In this situation, when the mobile communication terminal 93 approaches the communication area of the handover destination wireless base station 92, the superordinate apparatus 61 recognizes this situation and continues to transmit the IP packets 31 to be delivered to the mobile communication terminal 93 to the handover origin wireless base station 91, and also starts transmitting the IP packets 31 to the handover destination wireless base station 92. At this time, to ensure that the handover origin wireless base station 91 is synchronized with the handover destination wireless base station 92, the superordinate apparatus 61 transmits a control signal 71 indicating a start of the synchronization to the handover origin wireless base station 91. Having received the control signal indicating the start of the synchronization, the handover origin wireless base station 91 performs an erasure correction encoding process on the data that had been stored therein before the point in time at which the control signal 71 was received, in the same manner as in the first embodiment where the process is performed until a predetermined period of time has elapsed after the start of the storing of the data. The handover origin wireless base station 91 thus completes the transmission.
Having started to receive the IP packets 31 to be delivered to the mobile communication terminal 93, the handover destination wireless base station 92 receives, from the superordinate apparatus 61, a control signal 72 indicating a start of the transmission and starts transmitting erasure correction coded packets to the mobile communication terminal 93. In this situation, by having an arrangement in which the starting number of the sequence numbers assigned to the erasure correction coded packets is different from the starting number for the handover origin wireless base station 91, it is possible to perform the encoding process in such a manner that the sequence numbers do not overlap between these wireless base stations.
The mobile communication terminal 93 receives the erasure correction coded packets from both the handover origin wireless base station 91 and the handover destination wireless base station 92, performs an erasure correction decoding process, and also puts the received sets of packets together. When a decoding process has successfully been performed, the mobile communication terminal 93 generates the delivery confirmation signals 42 and 44 and transmits the generated delivery confirmation signals 42 and 44 to the handover origin wireless base station 91 and the handover destination wireless base station 92, respectively.
In the case where, after communicating with both of the wireless base stations at the same time as described above, the mobile communication terminal 93 moves to the communication area of the handover destination wireless base station 92, the superordinate apparatus 61 transmits the control signal 71 indicating a stop of the transmission to the handover origin wireless base station 91. Based on the received control signal 71, the handover origin wireless base station 91 stops transmitting the erasure correction coded packets to the mobile communication terminal 93 and clears the buffer of the stored data to be delivered to the mobile communication terminal 93.
In contrast, in the case where, after communicating with both of the wireless base stations at the same time as described above, the mobile communication terminal 93 returns to the communication area of the handover origin wireless base station 91, the subordinate apparatus 61 transmits the control signal 72 indicating a stop of the transmission to the handover destination wireless base station 92. As a result, the data stored in the handover origin wireless base station 91 is cleared from the buffer, and also, communication between the mobile communication terminal 93 and the handover origin wireless base station 91 is started.
As explained above, in the communication system according to the fourth embodiment, while the ARQ for the wireless section is terminated by one of the wireless base stations, the communication is handed over to the other wireless base station, without having to hand over the status in which the ARQ is performed from the one of the wireless base stations to the other. In addition, when the communication is handed over, control is exercised so that the transmission data signal to be delivered to the mobile communication terminal is transmitted from both of the base station apparatuses. Thus, it is possible to achieve a diversity effect between the wireless base stations, and also, it is possible to prevent or inhibit degradation of the user throughput related to the handovers.
Next, a communication system according to a fifth embodiment of the present invention will be explained, with reference to the each of
In the explanation of the fifth embodiment below, an example will be used in which the transmission data to be transmitted from the transmission-side communication apparatus to the reception-side communication apparatus is Medium Access Control Service Data Units (MAC_SDUs); however, the transmitted data may be Medium Access Control Protocol Data Units (MAC_PDUs) or other data in a superordinate layer.
In the first through the fourth embodiments, the configuration of the transmission data to be transmitted to the reception-side communication apparatus and the transmission controlling process were explained. However, according to the IEEE 802.16 standard, the term “concatenation” is used to refer to a process of generating desired transmission data from a plurality of pieces of data. Thus, in the explanation below, the term “concatenation” will be used.
Next, the concatenations in the communication system according to the fifth embodiment will be explained, with reference to
In
Further, as shown in
Another arrangement is acceptable in which each of the MAC_SDUs above contains, in a mixed manner, pieces of data corresponding to a plurality of connections, as the pieces of data to be concatenated, for example, in a BS-RS connection or an RS-RS connection.
Next, a flow in the data processing to be performed on the transmission side will be explained, with reference to
As explained also in the first embodiment, in the communication system according to the fifth embodiment, the predetermined number (i.e., K) indicating how many packets are included in each information packet group (RDT_SDU) and the maximum length Lmax of each information packet group are determined in a one-to-one correspondence. In this situation, in the case where a predetermined amount of data having a length called Basic SDU Length (BSL) has been stored within a predetermined period of time, an information packet group that has the length BSL defined in the following expression is generated:
BSL=the packet length L max×the predetermined number K indicating the number of packets (1)
The BSL defined in the expression above includes one or more padding bits used for adjusting the packet length.
On the other hand, in the case where the predetermined amount of data has not been stored within the predetermined period of time, an arrangement is acceptable in which the information packet group is generated by using the stored data and adding required padding bits, as necessary. By performing this process, it is possible to reduce the delay amount related to the storing of the data.
Also, in the example above, the predetermined number K indicating the number of packets and the packet length Lmax are the values that are determined for the system in a one-to-one correspondence. Alternatively, however, it is acceptable to use values that are determined by the system when the connection is established.
Returning to the description of
CP_Size=Ceiling(CL/K) (2)
In the expression above, Ceiling(a) denotes a ceiling function, which is a function that defines, with respect to a real number “a”, a smallest integer that is equal to or larger than the real number “a”.
Further, it is possible to express the length of the padding bits (i.e., Padding_Length) to be inserted into the RDT_SDU, by using the following expression:
Padding_Length=(CP_Size×K)−CL (3)
At the following step in the process, a packet (sometimes referred to as an information packet (SP) so as to be distinguished from a redundancy packet “PP”) that has a size expressed by CP_Size and on which an erasure correction encoding process has been performed based on the data in the RDT_SDU (including the padding bits) is generated. It is possible to adjust, in an arbitrary manner, the number and the size of the generated erasure correction coded packets. For example, as explained in the description of the first embodiment, it is acceptable to increase or decrease the number of erasure correction coded packets corresponding to the coding rate of “1”, without changing the packet size of each of the erasure correction coded packets. Alternatively, it is also acceptable to increase or decrease the packet size of each of the erasure correction coded packets, without changing the number erasure correction coded packets corresponding to the coding rate of “1”.
It is preferable to have an arrangement in which the number of erasure correction coded packets to be transmitted first is determined by adding as many redundancy packets (PPs) as a constant number a to the information packets (SPs). The constant number a may be a value that is configured for the system in a one-to-one correspondence or may be determined according to judgment based on transmission path information.
After the erasure correction coded packets have been generated, a CRC code is appended to one or more of the erasure correction coded packets. Also, a packet (called a “fragment”) is generated by connecting together a plurality of erasure correction coded packets to which the CRC code has been appended, according to the scheduled amount. Appended to the set of packets (i.e., the fragment) are a General_MAC header and a fragmentation sub-header/packing sub-header that are defined in the standards described in the Non-patent Documents listed above. A MAC_PDU to which, in addition to these sub-headers, an RDT sub-header that is newly defined according to the fifth embodiment is appended is generated. The generated MAC_PDU is transmitted to the reception side. It is also acceptable to append, as necessary, a CRC code for the header to the MAC_PDU. Also, another arrangement is acceptable in which a CRC code is appended in units that are variable for each connection or for each frame, depending on the state of the transmission path. As a result of this process, it is possible to balance the transmission efficiency and the error detection efficiency and to improve the throughput.
In the process described above, the MAC_PDU is generated by connecting together the units each of which is obtained by appending a CRC code to the one or more erasure correction coded packets. However, another arrangement is acceptable in which the MAC_PDU is generated by separating the erasure correction coded packets according to the partitions therein.
Next, the fragmentation sub-header and the packing sub-header that are defined in the IEEE 802.16 standard and modified so as to be applied to the fifth embodiment as well as the RDT sub-header that is newly defined so as to be applied to the fifth embodiment will be explained.
According to the fifth embodiment, a Group Sequence Number (GSN) is contained in the fragmentation sub-header and in the packing sub-header. By using the Group Sequence Number (GSN), even if the transmission-side communication apparatus transmits erasure correction coded packets having mutually different GSNs in parallel, the reception-side communication apparatus is able to recognize the fragmentation and the packing by using, as a unit, a group of erasure correction coded packets that is transmitted initially or additionally in correspondence with a certain GSN.
Also, because the RDT sub-header contains a packet number (i.e., a CPN) for the erasure correction encoding process and the size of each of the erasure correction coded packets, it is possible to identify the erasure correction coded packets on the reception side. Thus, it is also possible to change the size of each of the erasure correction coded packets depending on the transmission traffic and to improve the transmission efficiency. Another arrangement is acceptable in which the units in which the CRC code is appended to the erasure correction coded packets and the type of the CRC code may be configured for each system in a one-to-one correspondence or may be determined when the connection is established. Further, yet another arrangement is acceptable in which the units in which the CRC code is appended and the type of CRC code are notified to the reception side by using a control signal in an extended sub-header or the like. In addition, in the case where the size of each of the erasure correction coded packets is very small, it is acceptable to connect together a larger number of erasure correction coded packets than the predetermined number. However, it is preferable to append a CRC code at the end of every MAC_PDU.
On a CRC code appended to a header, it is possible to perform a process that is the same as the one performed on the CRC code appended to the erasure correction coded packets. More specifically, it is possible to select, as necessary, one of the following depending on the situation: the units in which the CRC code is appended and the type of the CRC code are (a) configured for each system in a one-to-one correspondence, (b) determined when the connection is established, and (c) notified to the reception side by using a control signal.
Next, the feedback information that is transmitted from the reception-side transmission apparatus to the transmission-side communication apparatus will be explained, with reference to
When the feedback information transmitted from the reception-side transmission apparatus has been received by the transmission-side communication apparatus, because delivery confirmations have already been completed for the Basic Group Sequence Number (BGSN: see
In consideration of a situation where the feedback information is lost, it is preferable to exercise transmission control by using a timer on the transmission side. The flow in this process is shown in
As for the number of times the additional transmission is performed, it is possible to specify the number of times so as to be a predetermined value that is equal to or larger than one, based on the QoS information of the connection, the transmission path information, or the like. Also, for example, in the case where the quality of the transmission path is good, needless to say, it is acceptable to specify the number of times the additional transmission is performed so as to be zero. Additionally, it is preferable to have an arrangement in which the number of coded packets transmitted from the transmission side is limited to the maximum number of coded packets that is determined based on the coding rate for the erasure correction code. For example, in the case where a code having a coding rate of ½ is used, twice as many erasure correction coded packets as the number of information packets are generated. Thus, when the transmission of all of the erasure correction coded packets have been completed, the additional transmission should be terminated.
In the case where the received erasure correction coded packets have successfully been decoded, the concatenation header (CH) in the reproduced information packet group (RDT_SDU) is analyzed. Also, the concatenations and the segmentations are cancelled so that the MAC_SDUs are reproduced. The reproduced MAC_SDUs are transferred to a superordinate layer either after order control is exercised in the case where it is necessary or the way they are in the case where order control is not necessary.
On the other hand, in the case where the decoding process has not successfully been performed, if the total number of received erasure correction coded packets has not reached the number of information packets, it is preferable to retain all the erasure correction coded packets or to perform concatenation and segmentation processes by using only the information packets that have successfully been reproduced and wait for erasure correction coded packets to be additionally transmitted. After that, when the erasure correction coded packets that are additionally transmitted have been received, the decoding procedure described above is performed on the received packets together with the erasure correction coded packets that have previously been received.
The feedback information may be transmitted to the transmission side either at regular intervals or when the decoding process has been performed.
The setting for the delivery confirmation completion (ACK) is started when the decoding for the corresponding GSN has been completed.
The setting for the delivery confirmation incompletion (NACK) is started when, for example, one of the following situation arises:
(1) Although a MAC_PDU corresponding to the last fragment has been received among the MAC_PDUs that have mutually the same GSN and are initially transmitted, the total number of received erasure correction coded packets has not reached the number of information packets;
(2) Even if the total number of erasure correction coded packets that have successfully been received in correspondence with a certain GSN has not reached the number of information packets, an erasure correction coded packet having a new GSN is to be received;
(3) In the case where the timer is activated after a MAC_PDU having a new GSN has been received, and the timer has expired before the conditions described above including the delivery confirmation completion are satisfied.
In
In the case where it is desired that the decoding process corresponding to the GSN should be terminated on the reception side before the purge timer (RDT_RX_PURGE) on the reception side expires when, for example, the additional transmission is no longer performed on the transmission side, it is acceptable to make a request to the reception side that the decoding process should be terminated by exchanging control signals.
In addition, in the case where it has been judged on the transmission side or on the reception side that the GSNs or the like are not synchronized, it is acceptable to reset various types of status and parameters used in the transmission control, by exchanging control signals.
Next, a communication system according to a sixth embodiment of the present invention will be explained, with reference to
The BS 301 has, as the layers that are subordinate to a convergence sub-layer 601, an Upper-MAC layer 611 used for forwarding MAC_PDUs to the MS 303, a Lower-MAC layer 621 used for forwarding MAC_PDUs to the RS 302, and a physical layer 631. The Upper-MAC layer 611 has an ARQ function 612, a MAC_PDU generating (i.e., framing) function 613, and a ciphering function 614. The Lower-MAC layer has a frame aggregation function 622 as well as an RDT_with_ECC function 623 and a MAC_PDU generating (i.e., framing) function 624 that are explained in the description of the fifth embodiment.
The RS 302 has: a MAC layer 651 that is used for forwarding MAC_PDUs to the BS 301 and processes superordinate data between the RS 302 and the MS 303 in a transparent manner; a physical layer 641 on the BS 301 side; and a physical layer 661 on the MS 303 side. Like the Lower-MAC layer 621 included in the BS 301, the MAC layer 651 has a frame aggregation function 652, an RDT_with_ECC function 653, and a MAC_PDU generating (i.e., framing) function 654.
The MS 303 has, as the layers that are subordinate to a convergence sub-layer 691, a MAC layer 681 used for forwarding MAC_PDUs to the BS 301, and a physical layer 671. Like the Upper-MAC layer 611 included in the BS 301, the MAC layer 681 has an ARQ function 682, a MAC_PDU generating function 683, and a ciphering function 684.
According to the sixth embodiment, the RS 302 does not need to have any key for the ciphering process. In addition, even if the RS 302 is introduced between the BS 301 and the MS 303, the BS 301 is able to use a conventional MAC layer as the Upper-MAC layer 611 without the need to modify it, while the Lower-MAC layer 621 is added to the configuration. Thus, an advantageous effect is achieved where there is no need to modify the MS 303.
As explained above, the communication system, the transmission-side communication apparatus and the reception-side communication apparatus according to the present invention are especially useful in the application to a communication system and the apparatuses included in the communication system that use the Automatic Repeat reQuest (ARQ) method in which the reception side automatically makes a request to the transmission side that transmission data should be re-transmitted.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2005-362277 | Dec 2005 | JP | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/JP2006/321825 | 11/1/2006 | WO | 00 | 5/1/2008 |
