Method and System for Automatic Optimization of Data Throughput Using Variable Packet Length and
Code Parameters
Field of the Invention
The present invention relates to data communication networks, and more particularly to optimizing data throughput in communication networks.
Background
The transmission medium and impairments to a communication channel due to noise and interference typically cause errors in data sent through communication networks. Where the channel error-rate has wide variations, use of forward error correction techniques can be inefficient. The inefficiency is brought about by the constant transmission of a large amount of redundancy so that heavily-errored patterns will not be interpreted incorrectly. The redundancy must always be transmitted even though the heavily-errored patterns may be rare. In addition, forward error correction algorithms have a tendency to require complex or computationally intensive decoders to determine which bits are incorrect.
Conventional automatic repeat request (ARQ) algorithms which packetize data, detect errors, and request retransmission of packets are implemented with less complexity than forward error correction methods but are still vulnerable to the design compromise of including large amounts of redundancy in order to avoid falsing on receive sequences containing large numbers of errors.
Thus, there is a need for a method and a communication system having efficient and highly reliable error-detection ihat automatically optimize data throughput of a channel.
Brief Descriptions of the Drawings
FIG. 1 is a block diagram of a communication system that operates in accordance with the method of the present invention.
FIG. 2 illustrates the format of the efficient, long packet mode and the robust, short packet mode.
FIG. 3 is a flow chart of an embodiment of steps in accordance with the method of the present invention.
FIG. 4 is a flow chart showing a particular embodiment of the method of the present invention wherein the input packet is in the efficient, long packet mode.
FIG. 5 is a flow chart showing another embodiment of the method of the present invention wherein the input packet is in the robust, short packet mode.
Detailed Description of a Preferred Embodiment
The present invention provides a method and a communication system having efficient and highly-reliable error-detection that automatically optimize data throughput of a channel whose error properties vary over time by selecting a length of a packet and a channel coding rate in a packet such that data throughput is maximized during periods when relatively few errors are found in packets and data redundancy is increased to provide a more robust system during periods when the number of errors of three particular
types of errors exceeds three respective predetermined threshold levels.
FIG. 1 , numeral 100, is a block diagram of a communication system that operates in accordance with the method of the present invention. The invention includes a communication system for maximizing channel efficiency in a data channel having varying transmission error conditions, wherein the communication system includes a transmitting unit (102) and a receiving unit (104), described more fully as follows. The transmitting unit (102) includes: a buffered data source (106), for providing data for transmitting; a selectable encoder (108), operably coupled to the buffered data source (106), for encoding the data into packets in accordance with a predetermined scheme using one of: an efficient channel encoder (110) and a robust channel encoder (112); a first transceiver (114), operably coupled to the selectable encoder (108), for transmitting the packets of data to the receiving unit (104) and for receiving error information from the receiving unit (104).
Buffered data source (106) provides a stream of data bits which may be grouped in information vectors U1 , U2, ..., UL, where L is a first predetermined integer, and the channel encoder (110) typically maps the vectors into channel vectors
V1 , V2 VL, or alternatively, buffered data source (106) provides a stream of data bits which may be grouped in information vectors u1 , u2, ..., us, where s is a second predetermined integer, and the channel encoder (112) then typically maps the vectors into channel vectors v1 , v2, ..., vs. In the preferred embodiment, L is selected to be greater than s.
The receiving unit (104) includes: a second transceiver (116), for receiving the transmitted data from the first
transceiver (114) and transmitting error information from a channel decoder (118) of the receiving unit (104); the channel decoder (118), operably coupled to the second transceiver (116), for decoding the received data and determining the error information wherein at least one of a-d: a) upon receiving a packet and determining that at least a predetermined number of codewords of the packet have errors, sending a signal for resetting the selectable encoder (108) of the transmitting unit (102) from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a second predetermined packet length and resetting the channel decoder (118) of the receiving unit (104) accordingly, b) upon requesting a predetermined number of retransmissions and receiving the same packet having errors, sending a signal for resetting the selectable encoder (108) of the transmitting unit (102) from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a second predetermined packet length and resetting the channel decoder (118) of the receiving unit (104) accordingly, c) upon receiving a packet having a codeword with an error count greater than a predetermined threshold, sending a signal for resetting the selectable encoder (108) of the transmitting unit (102) from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a second predetermined packet length and resetting the channel decoder (118) receiving unit (104) accordingly, and d) upon receiving a predetermined number of consecutive error-free packets, sending a signal for resetting the selectable encoder (108) of the transmitting unit (102) from a second predetermined channel coding rate to a first predetermined channel coding rate and changing a packet length from a second predetermined packet length to a first predetermined packet
length and resetting the channel decoder (118) of the receiving unit (104) accordingly; and a data sink (120), operably coupled "to the channel decoder (118), for receiving decoded received data. For (a)-(d), the second predetermined channel coding rate is less than the first predetermined channel coding rate and the second predetermined packet length is less than the first predetermined packet length.
Typically, for a 2.4 GHz propagation environment and a nominal one Megabit per second data transmission rate, the first predetermined channel coding rate is approximately 97 per cent and the second predetermined channel coding rate is approximately 70 per cent. For a 2.4 GHz propagation environment and a nominal one Megabit per second data transmission rate, the first predetermined packet length is generally in a range from 15 ms to 50 ms and the second predetermined packet length is in a range from 1 ms to 5 ms.
FIG. 2, numeral 200, illustrates the format of the efficient, long packet mode and the robust, short packet mode. The diagram depicts allocation of time from the perspective of the selectable encoder (108) of the transmitting unit (102). Transmitted packets include transmitter overhead time, TxOv, which is fixed and independent of packet length and channel coding rate. Also independent of packet length and channel coding rate are the time, ACK, for first transceiver (114) to receive acknowledgment from second transceiver (116). A third time period, propagation time, is dependent on physical separation of transmitting unit (102) and receiving unit (104) but like TxOv and ACK independent of packet length and channel coding rate.
In the efficient, long-packet mode (202) packets include L equal-length codewords, V1 through VL, each of which is comprised of information data bits requiring time, k1 , to
transmit and parity (redundancy) bits requiring time, p1 , to transmit.
In the robust, short-packet mode (204) packets include equal-length codewords, v1 through vs, each of which is comprised of information data bits requiring time, k2, to transmit and parity (redundancy) bits requiring time, p2, to transmit with second channel coding rate, k2/(k2+p2) less than first channel coding rate, k1/(k1 +p1 ).
Typically, for a 2.4 GHz propagation environment and a nominal one Megabit per second data transmission rate, (TxOv + propagation time + ACK) is in the range of 150 microseconds to 400 microseconds, codeword lengths k1+p1 = k2+p2 = 1 millisecond, k1 = 970 microseconds, k2 = 700 microseconds; L = 15 to 50 codewords, and s = 1 to 5 codewords.
FIG. 3, numeral 300, is a flow chart of an embodiment of steps in accordance with the method of the present invention. The method provides for maximizing channel efficiency in a data channel having varying transmission error conditions and includes the steps of at least one of a-d: a) upon receiving a packet and determining that at least a predetermined number of codewords of the packet have errors, sending a signal for resetting a transmitter from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a second predetermined packet length and resetting the receiver accordingly (304); b) upon requesting a predetermined number of retransmissions and receiving the same packet having errors, sending a signal for resetting the transmitter from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a
second predetermined packet length and resetting the receiver accordingly (306); c) upon receiving a packet having a "codeword with an error count greater than a predetermined threshold, sending a signal for resetting the transmitter from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a second predetermined packet length and resetting the receiver accordingly (308); and d) upon receiving a predetermined number of consecutive error-free packets, sending a signal for resetting the transmitter from a second predetermined channel coding rate to a first predetermined channel coding rate and changing a packet length from a second predetermined packet length to a first predetermined packet length and resetting the receiver accordingly (310), wherein, for (a)-(d) the second predetermined channel coding rate is less than the first predetermined channel coding rate and the second predetermined packet length is less than the first predetermined packet length (302).
When the method of FIG. 3 is implemented in a 2.4 GHz propagation environment using a nominal one Megabit per second data transmission rate, the first predetermined channel coding rate is typically approximately 97 per cent and the second predetermined channel coding rate is approximately 70 per cent. When the method of FIG. 3 is implemented in a 2.4 GHz propagation environment using a nominal one Megabit per second data transmission rate, the first predetermined packet length is generally in a range from 15 ms to 50 ms and the second predetermined packet length is in a range from 1 ms to 5 ms.
FIG. 4, numeral 400, is a flow chart showing a particular embodiment of the method of the present invention wherein the communication system is in the efficient, long packet
mode. The method is used for maximizing channel efficiency for transmitting packets in a data channel having varying "transmission error conditions and includes, for a received packet in the efficient mode, the steps of: a) clearing a retransmission flag for the received packet (401 ); b) decoding next available codeword in the packet and determining an error count of the codeword (402); c) determining whether the error count is > 0 (404); d) where the error count is > 0 and where the error count is < the predetermined error threshold (406), incrementing the codeword error count by 1 and setting the retransmission flag (408), determining whether the codeword error count is ≥ a predetermined codeword error threshold (409) , and where the codeword error count is > the predetermined codeword error threshold, waiting for an- end of the entire packet (410) and sending a message indicating that the packet failed and transitioning the receiver and a transmitter from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a second predetermined packet length, wherein, for the second predetermined channel coding rate is lower than the first predetermined channel coding rate and the second predetermined packet length is less than the first predetermined packet length (412) and transitioning to a robust mode; e) where the error count is > 0 and where the error count is > the predetermined error threshold, waiting for an end of an entire packet (410) and sending a message indicating that the packet failed and transitioning the receiver and a transmitter from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a second predetermined packet length, wherein, for the second predetermined channel coding rate is lower than the first predetermined channel coding rate and the second
predetermined packet length is less than the first predetermined packet length (412) and transitioning to a τobust mode, and where the codeword error count is < the predetermined codeword error threshold, upon receiving the entire packet, setting a retransmission flag (414), and where one of: the error count is = 0 (determined in 404) and the retransmission flag is set (414), determining whether a last codeword of the packet has been received (416), and where an intermediate codeword of the packet has been received, recycling to decoding a next available codeword (402), and where a last codeword of the packet has been received, checking whether the retransmission flag is set (418), and where the retransmission flag has been set, incrementing a retransmission count (420), determining whether a retransmission count is > a predetermined retransmission threshold (422), and where the predetermined retransmission count is > the predetermined retransmission threshold, sending a message indicating that the packet failed and transitioning the receiver and a transmitter from a first predetermined channel coding rate to a second predetermined channel coding rate and changing a packet length from a first predetermined packet length to a second predetermined packet length, wherein the second predetermined channel coding rate is less than the first predetermined channel coding rate and the second predetermined packet length is less than the first predetermined packet length (412) and transitioning to the robust mode, and where the predetermined retransmission count is < the predetermined retransmission threshold, sending the message indicating that the transmission failed and that the packet should be retransmitted, and where the retransmission flag is clear, resetting a transmission count, sending the message indicating that the transmission is successful (426) and recycling to clearing a retransmission flag (401 ).
FIG. 5, numeral 500, is a flow chart showing another -embodiment of the method of the present invention wherein the input packet is in the robust, short packet mode. The method provides for maximizing channel efficiency for transmitting packets in a data channel having varying transmission error conditions, comprising the steps of, for a received packet: a) clearing an error flag of the received packet (502); b) decoding a next available codeword of the received packet (504); c) determining whether there are error(s) in the codeword (506); d) where there is at least a first error in the codeword, setting the error flag (508) and determining whether the codeword is a last codeword of the packet (510); e) where the codeword is error-free, determining whether the codeword is a last codeword of the packet (510); f) where the codeword is an intermediate codeword of the packet, recycling to decoding a next codeword (504); g) where the codeword is a last codeword of the packet, checking whether the error flag is set (512); g) where the error flag is set, resetting an error-free count (514), sending the message indicating that the transmission failed and that the packet should be retransmitted (516, 524) and recycling to clearing the error flag (502); h) where the error flag is clear, incrementing the error-free count (518) and determining whether the error free count is > a predetermined threshold (520); i) where the error free count is < a predetermined threshold, sending the message indicating that the transmission is successful (522, 524) and recycling to clearing the error flag (502); j) where the error free count is > a predetermined threshold, sending a message indicating that the packet is successful (526, 528) and transitioning the receiver and a transmitter from a second predetermined channel coding rate to a first predetermined channel coding rate and changing a packet length from a second predetermined packet length to a first predetermined packet length, wherein,
for the first predetermined channel coding rate is greater than the second predetermined channel coding rate and the first -predetermined packet length is greater than the second predetermined packet length.
Although exemplary embodiments are described above, it will be obvious to those skilled in the art that many alterations and modifications may be made without departing from the invention. Accordingly, it is intended that all such alterations and modifications be included within the spirit and scope of the invention as defined in the appended claims.