Is a mark made by a sender and recognised easily by the receiver as belonging to the
|
journal article Show
MIS Quarterly Vol. 29, No. 2, Special Issue on Information Technologies and Knowledge Management (Jun., 2005) , pp. 197-219 (23 pages) Published By: Management Information Systems Research Center, University of Minnesota https://doi.org/10.2307/25148677 https://www.jstor.org/stable/25148677 Abstract The shift to more distributed forms of organizations and the prevalence of interorganizational relationships have led to an increase in the transfer of knowledge between parties with asymmetric and incomplete information about each other. Because of this asymmetry and incompleteness, parties seeking knowledge may not be able to identify qualified knowledge providers, and the appropriate experts may fail to be motivated to engage in knowledge transfer. We propose a sender-receiver framework for studying knowledge transfer under asymmetric and/or incomplete information. We outline four types of information structures for knowledge transfer, and focus on the sender-advantage asymmetric information structure and the symmetric incomplete information structure. We develop formal game-theoretical models, show how information incompleteness and asymmetry may negatively influence knowledge transfer, and propose solutions to alleviate these negative impacts. Implications for knowledge transfer research and practice are also discussed. Journal Information The editorial objective of the MIS Quarterly is the enhancement and communication of knowledge concerning the development of IT-based services, the management of IT resources, and the use, impact, and economics of IT with managerial, organizational, and societal implications. Professional issues affecting the IS field as a whole are also in the purview of the journal. Publisher Information Established in 1968, the University of Minnesota Management Information Systems Research Center promotes research in MIS topics by bridging the gap between the corporate and academic MIS worlds through the events in the MISRC Associates Program. Sassan Ahmadi, in
Mobile WiMAX, 2011 Automatic Repeat Request (ARQ) is an error-control mechanism for data transmission which uses acknowledgements
(or negative acknowledgements) and timeouts to achieve reliable data transmission over an unreliable communication link. In an ARQ scheme, the receiver uses an error detection code, typically a Cyclic Redundancy Check (CRC), to detect whether the received packet is in error. If no error is detected in the received data, the transmitter is notified by sending a positive acknowledgement. If an error is detected, the receiver discards the packet and sends a negative acknowledgement to the
transmitter, and requests a re-transmission. An Acknowledgement (ACK) or Negative Acknowledgement (NACK) is a short message sent by the receiver to the transmitter to indicate whether it has correctly or incorrectly received a data packet, respectively. Timeout is a predetermined time interval after the sender sends the packet; if the sender does not receive an acknowledgement before the timeout, it usually re-transmits the packet until it receives an acknowledgement or exceeds a predefined
number of re-transmissions. There are three types of ARQ protocol including [4]: Stop-and-Wait ARQ is the basic form of ARQ protocol where the sender sends one packet at a time and then waits for an ACK or NACK signal from the receiver before sending the same or a new packet. The receiver sends an ACK signal following receipt of a good packet. If the ACK does not reach the sender before the timeout, the sender re-sends
the same packet. Go-Back-N ARQ is a form of ARQ protocol in which the sender continuously sends a number of packets (determined by the duration of transmission window) without receiving an ACK signal from the receiver. The receiver process keeps track of the sequence number of the next packet it expects to receive, and sends the sequence number with every ACK it sends. The receiver will ignore any packet that does not have the exact
sequence number it expects whether that packet is a duplicate of a packet it has already acknowledged or a packet with a sequence number higher than the one expected. Once the sender has sequentially sent all the packets in its transmission window, it will check whether all of the packets are acknowledged and will resume sequential transmission of the packets starting with the next sequence number to the one that was last acknowledged. Selective
Repeat ARQ is a form of the ARQ protocol for transmission and acknowledgement of packets or fragments of a packet where the sending process continues to send a number of packets specified by a window size even after a packet is lost. Unlike Go-Back-N ARQ, the receiving process will continue to accept and acknowledge packets sent after an initial error. The receiver process keeps track of the sequence number of the earliest packet it has not received, and sends that number with every ACK it
sends. If a packet from the sender does not reach the receiver, the sender continues to send subsequent packets until it has emptied its window. The receiver continues to fill its receiving window with the subsequent packets, replying each time with an ACK containing the sequence number of the earliest missing packet. Once the sender has sent all the packets in its window, it re-sends the packet number given by the ACKs, and then continues where it stopped. Figure 7-2 provides an example illustration of the ARQ variants. It shows how efficiently different ARQ schemes utilize the communication channel, and use of the ARQ buffer in the transmitter and receiver. The major advantage of ARQ over Forward Error Correction (FEC) schemes is that error detection requires much simpler decoding mechanisms and much less redundancy than error correction. Furthermore, ARQ is
adaptive, in the sense that information is re-transmitted only when errors occur. On the other hand, FEC may be desirable instead of, or in addition to, error detection, for any of the following reasons: (1) a feedback channel is not available or ARQ delay is not tolerable; (2) the re-transmission scheme is not conveniently implemented; and (3) the expected number of errors without correction would require excessive re-transmissions. FIGURE 7-2. An example illustration of different ARQ schemes [4] The ARQ can be used for each connection between the MS and BS. Since the use of ARQ may increase the latency due to more reliability requirements, the ARQ mechanism is usually disabled for delay-sensitive applications, such as VoIP or interactive gaming. If the ARQ mechanism is enabled, the ARQ parameters are specified and negotiated during connection set-up. A connection does not contain a mix of ARQ and non-ARQ traffic. The scope of a specific instance of ARQ is limited to one uni-directional flow. 7.1.2 IEEE 802.16m ARQ MechanismAn ARQ block is generated from one or multiple MAC SDUs or MAC SDU fragments corresponding to the same flow. The ARQ blocks can be variable in size. An ARQ block is constructed by fragmenting MAC SDU or packing MAC SDUs and/or MAC SDU fragments. The fragmentation or packing information for the ARQ block is included in the extended header within the MAC PDU. When a MAC PDU is generated for transmission, the MAC PDU may contain one or more ARQ blocks. If the MAC PDU contains traffic from a single connection, the MAC PDU itself will be a single ARQ block. If information from multiple ARQ connections is multiplexed into one MAC PDU, the MAC PDU contains multiple ARQ blocks. The number of ARQ blocks in a MAC PDU is equal to the number of ARQ connections multiplexed in the MAC PDU. The ARQ blocks of a connection are sequentially numbered. The ARQ block Sequence Number (SN) is included in the MAC PDU using FPEH or MEH headers. The original MAC SDU ordering is maintained [2]. In the legacy system, the size of the ARQ blocks is fixed and the length of the ARQ blocks is specified by the serving BS for each connection and signaled through MAC management messages [1]. In that case, if the length of the MAC SDU is not an integer multiple of ARQ block size, the last ARQ block may be padded. The MAC SDU partitioning into ARQ blocks remains in effect until all ARQ blocks are received and acknowledged by the receiver [1]. As shown in Figure 7-3, if the initial transmission of an ARQ block fails, a re-transmission is scheduled with or without rearrangement. In the case of ARQ block re-transmission without rearrangement, the MAC PDU contains the same ARQ block and corresponding fragmentation and packing information which was used in the initial transmission. In the case of ARQ block re-transmission with rearrangement, a single ARQ block may be fragmented into a sequence of multiple ARQ sub-blocks. A MAC PDU payload is constructed from one or more ARQ sub-blocks. The ARQ sub-blocks are sequentially numbered using ARQ block SUB_SN (SSN). The size of an ARQ sub-block is defined by ARQ_SUB_BLOCK_SIZE, which is fixed. The ARQ sub-block is maintained during re-transmission. Figure 7-3 illustrates ARQ block initial transmission and re-transmissions. Two options for re-transmission are shown, i.e., with and without rearrangements of the failed ARQ blocks. FIGURE 7-3. ARQ block initial transmission and re-transmission [2] The ARQ feedback Information Element (IE) is defined for the receiver to indicate the reception status of an ARQ block (initial transmission) and an ARQ sub-block. The ARQ feedback IE is transported either as part of an extended header (piggybacked) within a MAC PDU or a standalone MAC control message. The ARQ feedback IE supports cumulative and selective ACK. In cumulative ACK, ARQ SN or ARQ SUB_SN are reported to indicate successful reception. In selective ACK, each bit of the ACK MAP indicates the error or success of ARQ blocks. The transmitter can request ARQ feedback poll to update the reception status of the transmitted ARQ blocks. In the downlink, an ABS may assign unsolicited bandwidth for the MS to send ARQ feedback information. The receiver sends ARQ feedback IE when these three conditions are met: (1) ARQ feedback polling request is received from the transmitter; (2) an ARQ block has been missing for a predetermined period; and (3) an ARQ discard message is received from the transmitter. The following parameters characterize an ARQ operation [2]: •ARQ_SN_MODULUS: the number of unique ARQ sequence values equal to 1024; •ARQ_WINDOW_SIZE: the maximum number of ARQ blocks with consecutive Block Sequence Number (BSN) in the sliding window of ARQ blocks that is managed by the receiver and the transmitter; •ARQ_BLOCK_LIFETIME: the maximum time interval an ARQ block is managed by the transmitter ARQ state machine, once initial transmission of the block has occurred. After expiring ARQ_BLOCK_LIFETIME, the corresponding ARQ block is discarded in the ARQ window; •ARQ_RX_PURGE_TIMEOUT: the time interval the receiver waits after successful reception of a block that does not result in advancement of ARQ_RX_WINDOW_START, before advancing ARQ_RX_WINDOW_START; •MAX_ ARQ_BUFFER_SIZE: the maximum size of the buffer in bytes that the MS is able to allocate for the ARQ connection; •ARQ_SYNC_LOSS_TIMEOUT: the maximum time interval ARQ_TX_WINDOW_START or ARQ_RX_WINDOW_START is allowed to remain at the same value before declaring a loss of synchronization of the sender and receiver state machines when data transfer is known to be active; •ARQ_ERROR_DETECTION_TIMEOUT: the time interval after which the ARQ block is declared as erroneous. It is used to reorder ARQ blocks that have not arrived in order due to HARQ re-transmission. 7.1.3 ARQ State MachineA finite-state machine can be used to model the behavior of the ARQ protocol in the transmitter and receiver sides. Each ARQ-enabled connection has an independent ARQ state machine. The ARQ state transitions are based on the status of ARQ blocks, rather than ARQ sub-blocks. Based on this model, an ARQ block may be in one of the following six states: NOT-SENT; OUTSTANDING; WAITING-FOR-RE-TRANSMISSION; DISCARD; REARRANGEMENT; or DONE. The ARQ state machine in the transmitter shown in Figure 7-4 is similar to that specified in the IEEE Standard 802.16-2009 [1]. The transmitter constructs each ARQ block using the fragmentation and packing rules. Each NOT-SENT ARQ block forms a MAC PDU and is assigned the value of the next ARQ block sequence number (ARQ_TX_NEXT_SN), which is then increased by one. The ARQ state machine variables are defined as follows:  FIGURE 7-4. ARQ state machine [2] •ARQ_TX_WINDOW_START is the lower edge of ARQ window in the transmitter; •ARQ_TX_NEXT_SN denotes the lowest ARQ sequence number of the next ARQ block to be sent by the transmitter; •ARQ_RX_WINODW_START is the lower edge of the ARQ window at the receiver; •ARQ_RX_HIGHEST_SN denotes the highest sequence number of the ARQ block received incremented by one. As shown in Figure 7-5, each ARQ block in the transmission buffer starts in NOT-SENT state before it is transmitted. When an ARQ block is initially transmitted, the ARQ_BLOCK_LIFETIME timer is set and the ARQ block transitions from NOT-SENT state to OUTSTANDING state. While an ARQ block is in OUTSTANDING state, the transmitter waits for an acknowledgement. If an acknowledgement is received, the ARQ block state transits to the DONE state. If a negative acknowledgement is received, the ARQ block state is changed to the WAITING-FOR-RE-TRANSMISSION state. When the ARQ_BLOCK_LIFETIME period expires, the ARQ block state is changed to the DISCARD state. While an ARQ block is in the WAITING-FOR-RE-TRANSMISSION state, the transmitter prepares for ARQ block re-transmission. If the ARQ block is re-transmitted, the ARQ block state is changed to OUTSTANDING. If the ARQ_BLOCK_LIFETIME period expires, the ARQ block state transits to the discard state. While the ARQ block is in the DISCARD state, the transmitter sends a discarded message and waits for acknowledgement from the receiver. If an acknowledgement for the ARQ block corresponding to the discarded message arrives, the ARQ block state transitions to the DONE state. When an ARQ block is in the DONE state, the transmitter flushes the ARQ block, and resets the timers and state variables associated with the flushed ARQ block.
FIGURE 7-5. ARQ block reconstruction at the receiver When a MAC PDU is received, the receiver examines the extended header and obtains the ARQ block information. Once the receiver identifies the ARQ block sequence number and the corresponding ARQ block in the MAC PDU, the receiver state machine adds this ARQ block to the list of blocks to be acknowledged. The state machine checks whether the block sequence number falls within the ARQ window range. If sequence number is not valid, the receiver discards the corresponding ARQ block; otherwise, if the corresponding ARQ block is already received, the state machine resets ARQ_RX_PURGE_TIMEMOUT timer and discards the ARQ block. If the received ARQ block is valid and not duplicated, the receiver state machine updates the ARQ state, as shown in Figure 7-5. The MAC SDUs are reconstructed from the received ARQ blocks and are sequentially delivered to the upper layers. The transmitter or receiver can reset the ARQ state machine, if necessary. When an ARQ reset error occurs during the ARQ reset procedure, the BS or MS may reinitialize the MAC procedures. When an ARQ block arrives out of order, each ARQ block with an intermediate sequence number is declared as missing, and the ARQ_ERROR_DETECTION_TIMEOUT for every missing ARQ block is set. If the missing ARQ block does not arrive within the ARQ_ERROR_DETECTION_TIMEOUT, the receiver declares the corresponding ARQ block as an error. The receiver sends feedback corresponding to each ARQ block using the ARQ feedback IE. The ARQ feedback is sent under one of the following conditions: an ARQ feedback poll is received from the transmitter; the receiver detects an ARQ block error; or when a discarded message is received from the transmitter. If all ARQ blocks in the ARQ window are received correctly, the ARQ feedback IE contains a cumulative acknowledgement. If one or more ARQ blocks in the ARQ window are in error, the ARQ feedback IE contains a selective acknowledgement to indicate the error. If the transmitter or receiver declares an ARQ synchronization loss, the transmitter or receiver may initiate the ARQ reset procedure. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B9780123749642100074 NR URLLCOlof Liberg, ... Gustav Wikström, in Cellular Internet of Things (Second Edition), 2020 11.3.3.2 HARQHARQ is the protocol handling the default method for triggering retransmissions in the PHY layer. Compared to automatically repeating a message, as with downlink and uplink repetitions described above, the fact that retransmissions are conditionally triggered in HARQ ensures a much-improved resource efficiency. A retransmission is triggered when an ACK is not received in HARQ, and since we target high reliability, most of the packets will indeed result in an ACK, and a retransmission will therefore only rarely be needed. HARQ can thus be seen as a way to use a higher code rate first and then gradually lowering it to the needed level. Incremental redundancy using a circular buffer is used (as in LTE) for this purpose of combining retransmissions into one longer code. This is indeed very resource efficient, but has a drawback of longer latency and the dependency on receiving the feedback. In NR the HARQ feedback is transmitted in UCI in the uplink for downlink data, and implicitly through a DCI in the downlink for uplink data. The latter is only true for NACK: an unsuccessful uplink transmission results in an uplink grant DCI for a retransmission, while a received uplink transmission results in no feedback on downlink. In NR, there is no equivalent to the LTE PHICH channel giving A/N feedback for uplink data transmissions. The main reason for this is that synchronous HARQ operation with hard-coded timing is not supported in NR, and both uplink and downlink operate using asynchronous HARQ with indicated timing. If the TB is segmented into CBs the HARQ feedback and retransmission of the CBGs is handled separately by using multiple indication bits in the UCI and DCI, which can improve resource efficiency since only the erroneous parts of a transmission are resent. NR HARQ is fully dynamic, meaning that there are no fixed connections between transmission and feedback. For downlink data the UCI where to send HARQ A/N is identified from the DCI, and for uplink data a DCI for retransmission is identified from the HARQ process index and a new data indicator, similar to LTE. There are 16 HARQ processes to use for a device. With HARQ, we can trigger retransmission to ensure high reliability at low resource cost, but only as long as the chain of feedback is not broken. Therefore, we cannot set an arbitrarily high error rate target (using high code rate) for transmissions and rely on many retransmissions, since at some point the feedback may not be delivered (it can also take a long time). However, the reliability of HARQ NACK doesn't need to be on the same level as that of data since it will be rare that the data transmission fails, requiring a retransmission. For both reasons of latency and reliability there is therefore a limit on how many HARQ retransmissions we can use for URLLC data. The more we can reliably use, the better for efficiency, which for uplink data appears a solid choice, but for downlink data we then rely on safe delivery of UCI which is not obviously achievable when the UE is power limited. This is discussed further in Chapter 12. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B978008102902200011X Communicating pictures: delivery across networksDavid R. Bull, Fan Zhang, in Intelligent Image and Video Compression (Second Edition), 2021 11.4.3 Hybrid ARQ (HARQ)Hybrid ARQ (HARQ) is a commonly used tool in modern mobile broadband networks such as 4G and 5G. HARQ [9] combines FEC coding (error detection and correction) with ARQ. It suffers from some of the basic limitations of ARQ, but offers some efficiency gains. In HARQ, the data is encoded with an appropriate FEC code, but the parity bits are not automatically sent with the data. Only when an error is detected at the decoder are these additional parity bits transmitted. If the strength of the FEC code is sufficient to correct the error, then no further action is taken. If however this is not the case, then the system reverts to a full ARQ retransmission. Typically a system will alternate between data and FEC packets during retransmission. HARQ is a compromise between conventional ARQ and conventional FEC. It operates as well as ARQ in a clean channel and as good as FEC in a lossy channel. In practice, with multiple retransmissions of the same data, a receiver would store all transmissions and use these multiple copies in combination. This is often referred to as soft combining. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B9780128203538000207 EC-GSM-IoTOlof Liberg, ... Gustav Wikström, in Cellular Internet of Things (Second Edition), 2020 3.2.8.4 More efficient HARQHARQ type II was introduced with EGPRS and is also used for EC-GSM-IoT. More details are provided in Section 3.3.2.2 on the HARQ operation for EC-GSM-IoT, but one can note here that for uplink operation, the use of Fixed Uplink Allocation (FUA) for allocation/assignment of resources will (which is more elaborated upon in Section 3.3.2.1) allow the receiver to operate at a higher BLock Error Rate (BLER) on the uplink, because no detection of the sequence number is required. A higher BLER of the traffic channel will effectively increase coverage (as long as the targeted minimum throughput is reached). Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B9780081029022000030 Communications Satellite SystemsTakashi Iida, Hiromitsu Wakana, in Encyclopedia of Physical Science and Technology (Third Edition), 2003 II.D.3.a Automatic repeat requestIn ARQ systems, if a receiving terminal does not detect errors in received data, it sends an acknowledgment (ACK) signal to the transmitting terminal. If the receiving terminal does detect errors, it send a negative acknowledgment (NACK) signal to notify that the data block was not correctly received and to request retransmission of the same block. Stop-and-wait, go-back-N, and selective-repeat request are three basic methods of ARQ. For satellite communication systems where error-correcting codes may not work reliably because of low and variable link quality or high interference, ARQ can provide adequate robustness and required error performance. Because of the long delay in satellite links, however, a long enough interval must be specified for timeout to prevent premature timeouts. If the sender performs timeout too early, while the ACK signal is still on the way, it will send a duplicate. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B0122274105008826 5G NR in the unlicensed spectrumHao Lin, ... Kevin Lin, in 5G NR and Enhancements, 2022 18.5.2.2 Type-3 hybrid automatic repeat request acknowledgment codebookAs mentioned earlier, when a UE is configured with the Type-3 HARQ–ACK codebook, a 1-bit trigger requesting feedback of Type-3 HARQ–ACK codebook (one-shot HARQ–ACK feedback request) is included in DCI format 1_1 and the gNB can trigger the request for one-shot HARQ–ACK feedback from the UE by setting the trigger bit to 1 in the DCI. The DCI format 1_1 the gNB triggers for Type-3 HARQ–ACK codebook feedback from the UE could be a DCI that schedules a PDSCH reception or does not schedule a PDSCH reception for the UE. When the UE receives a trigger from the gNB requesting Type-3 HARQ–ACK codebook feedback in DCI format 1_1, the UE must generate and report a Type-3 HARQ–ACK codebook. The Type-3 HARQ–ACK codebook includes HARQ–ACK feedback information for all of the HARQ processes and for all of the configured cells in a PUCCH group. For Type-3 HARQ–ACK codebook, there can be two feedback scenarios: one is feedback of a Type-3 HARQ–ACK codebook with a new data indicator (NDI) information and the other one is without NDI information. The gNB can configure using higher-layer parameters whether or not the UE should include the NDI information when reporting the Type-3 HARQ–ACK codebook. The generation of Type-3 HARQ–ACK codebook for the two scenarios are described below. •Scenario 1: Type-3 HARQ–ACK codebook feedback with NDI information In NR communication, every transmission of a data Transport Block (TB) is associated with an NDI value. In the case of Type-3 HARQ–ACK codebook feedback with NDI information, the UE must report the NDI and HARQ–ACK information for the latest received TB(s) in each HARQ process. If the UE does not receive a TB for a HARQ process, then the UE sets the NDI to 0 and the HARQ–ACK information to NACK for this HARQ process. The order of information encoded in the Type-3 HARQ–ACK codebook is arranged in the following sequence: first in ascending order of code block group (CBG) or TB index, second in ascending order of HARQ process number, and finally in ascending order of cell index. For each data TB, HARQ–ACK information is placed before the NDI information. Assuming that only one HARQ–ACK information bit corresponds to one HARQ process, a UE is configured with cell 1 and cell 2 in a PUCCH group, 16 HARQ processes are configured for each cell, and the UE may be scheduled for PDSCH reception in cell 1 and cell 2. Furthermore, DCI format 1_1 includes a trigger field for requesting feedback of Type-3 HARQ–ACK codebook (represented by T). As shown in Fig. 18.51, in slot n the UE receives a DCI scheduling PDSCH1 in cell 1 and in the DCI it is indicated that HARQ=4, NDI=1, and the PUCCH resource for feeding back HARQ–ACK information for the scheduled PDSCH1 is in slot n+3 by setting K1=3. In the scheduling DCI, T is set to 0, meaning the report of Type-3 HARQ–ACK codebook is not requested. In addition, the UE also receives a DCI scheduling PDSCH2 in cell 2 and in the DCI it is indicated that HARQ=5, NDI=0, and the PUCCH resource for feeding back HARQ–ACK information for the scheduled PDSCH2 is in slot n+3 by setting K1=3. In the scheduling DCI this time, T is set to 0 again, meaning Type-3 HARQ–ACK codebook feedback is not requested. Figure 18.51. One-shot hybrid automatic repeat request acknowledgment feedback. In slot n+1, the UE receives a DCI scheduling PDSCH3 in cell 1 and in the DCI it is indicated that HARQ=8, NDI=0, and the PUCCH resource for feeding back HARQ–ACK information for the scheduled PDSCH3 is in slot n+3 by setting K1=2. In the scheduling DCI, T is set to 0, meaning the report of Type-3 HARQ–ACK codebook is not requested. In slot n+2, the UE receives a DCI scheduling PDSCH4 in cell 2 and in the DCI it is indicated that HARQ=9, NDI=1, and the PUCCH resource for feeding back HARQ–ACK information for the scheduled PDSCH4 is in slot n+3 by setting K1=1. In the scheduling DCI, T is set to 1, meaning the report of Type-3 HARQ–ACK codebook is requested this time. The Type-3 HARQ–ACK codebook generated by the UE for PUCCH 1 in slot n+3 includes NDI information for each TB. More specifically, the 9th and 10th bits correspond to the decoding result and NDI information for the scheduled PDSCH1. The 17th and 18th bits correspond to the decoding result and NDI information for the scheduled PDSCH3. The 43rd and 44th bits correspond to the decoding result and NDI information for the scheduled PDSCH2. The 51st and 52nd bits correspond to the decoding result and NDI information for the scheduled PDSCH4. The details of the Type-3 HARQ–ACK codebook for this example are shown in Fig. 18.52. Figure 18.52. Hybrid automatic repeat request acknowledgment codebook. •Scenario 2: Type-3 HARQ–ACK codebook feedback without NDI information For the feedback scenario of Type-3 HARQ–ACK codebook feedback without NDI information, the UE must report HARQ–ACK information for each HARQ process. When a UE does not receive a data TB for a particular HARQ process, then the UE sets the HARQ–ACK feedback information to NACK for this HARQ process. In addition, the UE must also reset the HARQ–ACK feedback information to NACK for a TB of a HARQ process if the UE has already reported ACK for this TB previously. The order of information encoded in Type-3 HARQ–ACK codebook is arranged in the following sequence: first in ascending order of CBG or TB index, second in ascending order of HARQ process number, and finally in ascending order of cell index. Considering the same example in Fig. 18.51, the Type-3 HARQ–ACK codebook generated by the UE for PUCCH 1 in slot n+3 does not include NDI information of each TB for this reporting scenario 2. More specifically, the 5th bit corresponds to the decoding result of PDSCH1, the 9th bit corresponds to the decoding result of PDSCH3, the 22nd bit corresponds to the decoding result of PDSCH2, and the 26th bit corresponds to the decoding result for PDSCH4. The details of Type-3 HARQ–ACK codebook for this example are shown in Fig. 18.53. Figure 18.53. Hybrid automatic repeat request acknowledgment codebook. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B9780323910606000180 Error Detecting and Correcting CodesPatrick Verlinde, in Encyclopedia of Information Systems, 2003 II.A. ARQ: Error Detection and RetransmissionThe principle here is to increase the redundancy of the binary sequence by adding binary symbols that allow for the detection of errors. Therefore the binary sequence is divided into consecutive blocks to which binary control symbols are added. When at reception an error is detected, one does not try to correct this error. Instead a retransmission of the erroneous block is requested. This supposes, of course, the use of an acknowledgment, or ACK. With this strategy different procedures can be used according to the type of transmission channel. The transmission channels can be classified according to the possible transmission directions. In this way one defines simplex, half-duplex, and full-duplex transmission channels. A simplex transmission channel allows communication in only one direction. A half-duplex transmission channel allows communication in either direction, but only one at a time, while a full-duplex transmission channel allows simultaneous communication in both directions. It should be clear that the ARQ strategy cannot be used with a simplex transmission channel, since this strategy depends on the sending of an ACK by the receiver. In fact, three different ARQ procedures are possible. In the first case the transmitter waits until reception of the ACK for the block that he just transmitted, before transmitting the next block. In this case a half-duplex transmission channel is sufficient. In the two other cases, there is a simultaneous transmission of information blocks by the transmitter and of ACKs (or not-ACKs) by the receiver. The latter arrive at the transmitter with a certain delay with respect to the moment of transmission of the blocks to which they are related. It must thus be possible to identify these blocks. The difference between the second and the third case resides in the retransmission which is executed in case of reception of a not-ACK: complete retransmission of all blocks which follow the erroneous received block in the second case or just a retransmission of the erroneous block in the third case. These last two cases do require a full-duplex transmission channel. The advantages of the ARQ method over the FEC method are as follows: •Simplicity of coding and decoding since the “correction” aspect is not taken into account in the code. The redundancy needed in this ARQ strategy is therefore smaller than in the case of FEC. •The correction by retransmission has an adaptatif character in the sense that the redundancy needed for the “correction” (here the retransmission) only needs to be introduced when an error really occurs. •The ARQ method allows for very small residual error percentages. The drawbacks are: •The need for at least a half-duplex transmission channel. •The delay between the moment of transmission of a block and the instant that block is received correctly is variable in time. This delay can become unacceptable if many successive retransmissions are needed. •The need for a buffer memory for the transmitter. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B0122272404000629 5G V2X
Zhenshan Zhao, ... Kevin Lin, in 5G NR and Enhancements, 2022 17.3.1 Sidelink HARQ feedbackIn LTE–V2X, only broadcast transmission is supported for direct sidelink communication among vehicle and pedestrian UEs. To ensure a basic reliability requirement for the sidelink communication is met, data messages are blindly retransmitted, even when an initial transmission is correctly decoded by a receiver UE. In sidelink blind retransmissions, the TX UE does not perform repeated transmission of the same data TB according to a HARQ feedback from the Receiver UE (RX UE), but autonomously retransmits the same sidelink data TB a certain number of times before it performs transmission for a new data TB. For advanced V2X use cases, the reliability requirement for direct sidelink communication became significantly more stringent. If blind retransmission is still used as the main mechanism to fulfill the reliability target, the required number of blind retransmissions for each data TB would be large and such a scheme would not be very spectral efficient in scenarios where channel condition between the TX and RX UE is good, communication range is short, or there is only a small number of RX UEs. Furthermore, transmission latency would likely be prolonged as well due to unnecessary/excessive retransmissions. Therefore in order to meet the higher transmission reliability as well as improve the spectral utilization efficiency and transmission latency, two sidelink HARQ feedback schemes were developed and introduced for NR–V2X, namely ACK/NACK feedback and NACK-only feedback. Overall, in sidelink HARQ feedback, the RX UE decodes PSCCH/PSSCH sent by a TX UE, and then feeds back HARQ information using PSFCH to the TX UE when the sidelink HARQ indicator is set to enabled in SCI of the received PSCCH. 17.3.1.1 Sidelink HARQ feedback schemeNR–V2X supports three transmission types: unicast, groupcast, and broadcast. Sidelink HARQ feedback is only applicable in unicast and groupcast transmissions, but not for broadcast. In the broadcast transmission mode, it is the same as LTE–V2X; a TX UE blindly performs retransmissions for multiple times (network configured or preconfigured for the RP) to ensure the target transmission reliability is achieved. In the unicast transmission mode, after a TX UE and a RX UE have established a unicast communication link (i.e., a PC5-RRC connection) and the TX UE has sent a sidelink data transmission intended for the RX UE, the RX UE transmits PSFCH in a corresponding slot and resource containing HARQ information to the TX UE according to the decoding result as shown in Fig. 17.20. For sidelink unicast transmission, only the ACK/NACK feedback scheme is supported. Figure 17.20. Sidelink feedback. In groupcast transmission mode, both sidelink HARQ feedback schemes of ACK/NACK feedback and NACK-only feedback are supported. The TX UE directly indicates the HARQ feedback scheme to be used by the receiver in the 2nd-stage SCI when transmitting the PSSCH to the RX UE. •In the NACK-only feedback scheme, if a RX UE failed to decode a received PSSCH, it feeds back a NACK in PSFCH. If another RX UE successfully decodes the PSSCH, it does not send back any HARQ information. For all other RX UEs that needs to send a NACK due to decoding failure, they will send the same HARQ information (only NACK) using the same PSFCH resource corresponding to the sidelink resource in which the PSSCH was received. This HARQ feedback scheme was originally designed for connection-less groupcast transmissions, where no specific communication group is established. As such, this HARQ feedback scheme is very suitable for a type of V2X groupcast communication that is based on a communication distance requirement, and only the UEs within an indicated distance range from the TX UE need to feedback sidelink HARQ information to the TX UE. UEs that are outside the communication distance range need not send sidelink HARQ feedback. This NACK-only HARQ feedback scheme was later during the NR–V2X development phase extended to support also connection-based groupcast transmissions where the number of members in the group and group member ID are known to the TX UE. Since there may be a limited number of PSFCH resources that can be (pre)configured in a RP, there may not always be enough PSFCH resources available for sensing ACK/NACK reports from every UE in a group, especially when the group size is large (e.g., more than 12). As such, the NACK-only feedback scheme can be indicated in the 2nd-stage SCI from the TX UE when transmitting sidelink data in the PSSCH. For the communication distance range groupcast transmissions, the concept of geographical zone is utilized to support the NACK-only sidelink HARQ feedback scheme. The basic mechanism behind the geographical zone concept is to divide the surface of the earth into different zones, where each zone is identified by a zone ID. For the NACK-only HARQ scheme, TX UE’s SCI carries a zone ID corresponding to the zone to which the TX UE belongs and a communication distance range. UEs that are in geographical zones within the distance range from the transmitter zone are required to provide NACK-only reports when there is failure to decode the PSSCH from the TX UE. Specifically, when a RX UE receives SCI format 1–B sent from a TX UE with communication distance range and zone ID indicated, it determines the distance to the TX UE according to the indicated zone ID and the zone where the RX UE is located which can be derived base on its own real-time geo-location (e.g., from receiving GNSS and/or cellular network signals). However, the RX UE only knows the zone ID of the TX UE and does not know the actual geographic location of the TX UE. Therefore the RX UE determines the distance to the TX UE according to its real-time geographic location and the zone center location of the TX UE. If the distance determined by the RX UE is less than or equal to the distance range indicated in the received SCI and the associated PSSCH is not successfully decoded, the RX UE is required to feedback a NACK in the PSFCH. If the associated PSSCH is successfully decoded, no HARQ information should be fed back. If the calculated distance separation between the TX and RX UEs is larger than the distance indicated by the communication distance requirement, no HARQ information should be fed back either. As shown in Fig 17.21 below, a TX UE transmits sidelink control and data (PSCCH/PSSCH) in Zone 4 and indicates its zone ID and distance range information in SCI format 1–B. When receiver UE 1 and UE 2 determined its distance to the TX UE is less than the indicated communication distance requirement, they will need to feed back a NACK if there is a failure to decode the transmitted PSSCH from the TX UE; otherwise, no sidelink HARQ information feedback is necessary from either UE. UE 3 determines its distance to the TX UE is larger than the communication distance range indicated and thus will transmit no sidelink HARQ feedback information to the TX UE nor it will attempt to decode the transmitted PSSCH from the TX UE (Fig. 17.21). Figure 17.21. HARQ feedback based on zone and communication distance requirement. •In the ACK/NACK-based HARQ feedback scheme, if a UE successfully decodes a received PSSCH and the sidelink HARQ feedback indicator is set to enabled in the scheduling SCI, then the UE is required to feedback an ACK; otherwise it will feedback a NACK if decoded unsuccessfully. This sidelink HARQ feedback scheme is generally suitable for connection-based sidelink groupcast and unicast communications. In connection-based groupcast communication, a group of UEs constitute a communication group, and each UE in the group is given a corresponding member ID. For example, as shown in Fig. 17.22, a communication group consists of 4 UEs (i.e., the group size is 4) and each UE is given a member ID such as ID#0, ID#1, ID#2, and ID#3. Furthermore, each UE is also given information on the number of members within the group and the member IDs of all group member UEs. When a member UE transmits PSCCH/PSSCH (i.e., all other member UEs in the group are RX UEs), each RX UE will determine whether to feedback an ACK or a NACK to the TX UE according to its PSSCH decoding results. In this case, each RX UE uses a different PSFCH resource to feedback its sidelink HARQ information, and sidelink HARQ feedback information from all RX UEs are FDM’ed and/or CDM’ed in PSFCH. Figure 17.22. Sidelink feedback for groupcast. 17.3.1.2 Sidelink HARQ feedback resource configurationIn the PSSCH RP configuration information, sidelink HARQ feedback transmission resources (i.e., PSFCH resources) can be network configured or preconfigured in RRC. The configuration parameters for sidelink HARQ feedback transmission resources include the following four components: •Sidelink HARQ feedback resource period: For a sidelink RP, PSFCH resources for sidelink HARQ feedback can be configured in every transmission slot. But since not every sidelink data transmission requires SL HARQ feedback from the receiver UE, an occurrence period (P) for PSFCH resources can be (pre)configured to reduce the overhead of sidelink HARQ feedback resources, where P can be 0, 1, 2, or 4, expressed in number of slots in the sidelink RP. P=0 means no PSFCH feedback resource is (pre)configured and all sidelink HARQ feedback is disabled in the sidelink RP. •Time interval: This is used to indicate a minimum time gap between the sidelink HARQ feedback resource and its corresponding PSSCH transmission resource, expressed in number of slots. •Frequency-domain resource set of sidelink HARQ feedback resources: This is used to indicate the position and number of Resource Blocks (RBs) that can be used to transmit the PSFCH in a RP. This parameter is indicated in the form of a bitmap, and each bit in the bitmap corresponds to a RB in the frequency domain. •Number of Cyclic Shift Pairs (CS pairs) in a RB: The sidelink HARQ feedback information is carried in a form of pseudo-random sequences. ACK and NACK information bits are represented by different sequences, which make up a CS pair. This parameter is used to indicate the number of CS pairs (i.e., the number of UEs that can be multiplexed by CDM in a RB). 17.3.1.3 Sidelink HARQ feedback resource determinationThe PSFCH transmission resource is determined according to the time-frequency position of its corresponding PSSCH resources used for transmission. In NR–V2X, the following two PSFCH resource determination options are supported. The specific option for determining theh PSFCH resource is configured according to higher layer signaling. •Option 1: PSFCH resource is determined by the first subchannel used for the corresponding PSSCH transmission. •Option 2: PSFCH resource is determined by all the subchannels occupied by the corresponding PSSCH transmission. For PSFCH resource determination Option 1, since PSFCH resources are determined only by the first subchannel occupied by the PSSCH (i.e., not dependent on the number of subchannels the PSSCH occupied), the total number of corresponding PSFCH feedback resources is fixed for each PSSCH transmission. For Option 2, the number of PSFCH transmission resources is determined according to the number of subchannels occupied by PSSCH. Therefore the more subchannels occupied by a PSSCH transmission, the more PSFCH transmission resources are available for sidelink HARQ feedback. As such, Option 2 is more suitable for scenarios where a large number of member UEs are involved in a sidelink groupcast communication that requires more sidelink HARQ feedback resources for the ACK/NACK feedback scheme. The corresponding PSFCH transmission resource sets RPRB,CSPSFCH that are available for multiplexing sidelink HARQ information can be determined according to the slot and subchannel(s) used for the PSSCH transmission. The index of PSFCH resources in the resource set is assigned first in the ascending order of the allocated PFSCH frequency RBs, then in the ascending order of configured CS pairs. To determine the exact resource for PSFCH transmission, the RX UE uses the following formula: (17.4)(PID+MID)modRPRB ,CSPSFCH where PID is the Source ID of the TX UE indicated in the received 2nd-stage SCI. For unicast and NACK-only groupcast sidelink HARQ feedback schemes, MID is set to zero. For the ACK/NACK groupcast sidelink HARQ feedback scheme, MID is the member ID of the RX UE provided by the UE higher layer. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B9780323910606000179 Joint Source-Channel Coding for Video CommunicationsFan Zhai, ... Aggelos K. Katsaggelos, in Handbook of Image and Video Processing (Second Edition), 2005 5.2.2 Joint Source Coding and Hybrid Forward Error Correction and Automatic Repeat ReQuestWhen considering the use of both FEC and ARQ, the channel coding parameter c includes the FEC rate chosen to protect each packet and the retransmission policy for each lost packet. Hybrid FEC/retransmission has been considered in [22], where a general cost-distortion framework is proposed to study several scenarios such as DiffServ (Differentiated Services), sender-driven retransmission, and receiver-driven retransmission. In [58], optimal error control is performed by jointly considering source coding with hybrid FEC and sender-driven application-layer selective retransmission. This study is carried out with the use of (5), with a sliding window scheme in which lost packets are selectively retransmitted according to a rate-distortion optimized policy. Simulations in [58] show that the performance advantage in using either FEC or selective retransmission depends on the packet loss rate and the round-trip time. In that work, the proposed hybrid FEC and selective retransmission approach is able to derive the benefits of both approaches by adapting the type of error control based on the channel characteristics. A receiver-driven hybrid FEC/pseudo-ARQ mechanism is proposed for Internet multimedia multicast in [33]. In that work, the sender multicasts all the source layers and all the channel layers (parity packets obtained by using RS coding similar to what we have discussed in the previous section) to separate multicast groups. Each user computes the optimal allocation of the available bit rate between source and channel layers based on its estimated channel band-width and packet loss probability, and joins the corresponding multicast group. This is achieved through a pseudo-ARQ system, in which the sender continuously transmits delayed parity packets to additional multicast group, and the receivers can join or leave a multicast group to retrieve the lost information up to a given delay bound. Such a system looks like ordinary ARQ to the receiver and an ordinary multicast to the sender. This can be characterized as JSCC with receiver feedback. More specifically, the optimal JSCC is obtained by solving (5) at the receiver side, where the source coding parameter is the number of source layers, and the channel coding parameter is the number of channel layers. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B9780121197926501248 Adaptive FEC-Based Error Control for Internet TelephonyJean-Chrysostome Bolot, ... Don Towsley, in Readings in Multimedia Computing and Networking, 2002 I. INTRODUCTIONThe transmission of real time audio, and especially of real time voice, over the Internet has been much in the news recently. Traditional voice carriers, so-called nextgen telcos, as well as manufacturers of gateways, phone-like appliances, and routers, have all become involved in some way or another with Internet telephony. Internet telephony is branded by the various parties as fitting anywhere between “the old-telco killer app” and “a toy for long distance lovers”. In any case, it is clear that the field of packet voice over the Internet has matured and that the basic building blocks are available [25], ranging from high quality codecs to standardized packetization and signaling protocols such as RTP [24], H.323 [11], or SIP [26]. Still, Internet telephony has often been dismissed as a “real” application because of the mediocre quality experienced by many users of Internet voice software. Audio quality problems are not so surprising because the current Internet provides users with a single class best effort service which does not promise anything in terms of performance guarantees. And indeed, measurements show persistent problems with audio quality caused by congestion in the network, and thus by the impact of traffic in the network on the streams of audio packets. In practice, this impact is felt via high loss rates, varying delay, etc. In the absence of network support to provide guarantees of quality (such as a maximum loss rate or a maximum delay) to users of audio tools, a promising approach to tackle the problems caused by varying loss rates, delays, or available bandwidth, is to use application level control mechanisms. These mechanisms adapt the behavior of the audio application so as to eliminate or at least minimize the impact of loss, jitter, etc, on the quality of the audio delivered to the destinations. Efficient playout adjustment mechanisms have been developed to minimize the impact of delay jitter [16]. Much recent effort has been devoted to developing mechanisms to minimize the impact of loss. Rate control mechanisms attempt to minimize the number of packets lost by ensuring that the rate at which audio packets are sent over a connection matches the capacity of the connection [5]. However, they typically do not prevent loss altogether. An error control, or loss recovery, mechanism is required if the number of lost audio packets is higher than that tolerated by the listener at the destination. Typical mechanisms fall into one of two classes [19]. Automatic Repeat Request (ARQ) mechanisms are closed-loop mechanisms based on the retransmission of the packets that were not received at the destination. Forward Error Correction (FEC) mechanisms are open-loop mechanisms based on the transmission of redundant information along with the original information so that some of the lost original data can be recovered from the redundant information. ARQ mechanisms are typically not acceptable for live audio applications over the Internet because they dramatically increase end to end latency1. FEC is an attractive alternative to ARQ for providing reliability without increasing latency. FEC schemes send redundant information along with the original information so that the lost original data can be recovered, at least in part, from the redundant information. There are two main issues with FEC. First, the potential of FEC mechanisms to recover from losses depends in large part on the characteristics of the packet loss process in the network. Indeed, FEC mechanisms are more effective when the average number of consecutively lost packets is small. Second, sending additional redundancy increases the probability of recovering lost packets, but it also increases the bandwidth requirements and thus the loss rate of the audio stream. This means that the FEC scheme must be coupled to a rate control scheme. Furthermore, the amount of redundant information used at any given point in time should also depend on the characteristics of the loss process at that time (it makes no sense to send redundant information when the channel is loss free), on the end to end delay constraints (destination typically have to wait longer to decode the FEC as more FEC information is used), on the quality of the redundant information, etc. The problem, then, becomes a constrained optimization problem, namely: given constraints of the rate control mechanisms (i.e. given a total rate at which the source can send), find the combination of main and redundant information which provides the destination with the best perceived audio quality. It is precisely the goal of this paper to formalize this problem, solve it, derive a practical algorithm, apply it to the FEC scheme recently standardized in the IETF [18], and evaluate the performance of the algorithm in realistic Internet environments using a real Internet audio/telephony tool. The paper is organized as follows. In Section II, we first briefly review recent results on the the loss process of audio packets in the Internet. We then describe a simple FEC scheme which uses these results to minimize an objective function (the loss rate after packet reconstruction) at the destination. However, that scheme turns out to have a number of drawbacks. We describe in Section III our main contribution, namely an adaptive algorithm for the IETF FEC scheme which incorporates the constraints of rate control and playout delay adjustment, which adapts to varying loss conditions, and which maximizes a subjective measure of quality (such as the perceived audio quality at a destination) as opposed to a measure such as the packet loss rate at a destination which does not reflect the quality perceived by the receiver. We present simulation and experimental results in Section IV to illustrate the performance of the algorithm. Read full chapter URL: https://www.sciencedirect.com/science/article/pii/B9781558606517501406 Is a security service where receiver ensure that message is received from Authorised sender?The correct answer is Nonrepudiation. Nonrepudiation prevents either sender or receiver from denying a transmitted message. Nonrepudiation ensures that no party can deny that it sent or received a message via encryption and/or digital signatures or approved some information.
Which of the following refers to the property that the receiver is guaranteed that the message really came from the real sender?Definition(s): Assurance that the sender of information is provided with proof of delivery and the recipient is provided with proof of the sender's identity, so neither can later deny having processed the information.
Which of the following ensures that data received was sent by the specified sender?Nonrepudiation provides proof of the origin, authenticity and integrity of data. It provides assurance to the sender that its message was delivered, as well as proof of the sender's identity to the recipient.
Which of the following means that the receiver is ensured that the message is coming from the intended sender not an imposter?(c) Message authentication means the receiver is ensured that the message is coming from the intended sender.
|
