Optimizing packet size to eliminate effects of reception nulls

Description Submit all comments and votes

BACKGROUND OF THE INVENTION

This invention relates generally to increasing the reliability of wireless communication systems and more specifically to a system for varying the size of message packets according to the speed of motion of a receiver.

Radio paging systems and other types of wireless message broadcast systems transmit messages to remote receiving devices. For example, U.S. Pat. No. 4,713,808 to Gaskill et al. (Gaskill) describes a time division multiplexed (TDM) data protocol where pager messages are queued into 13.6 millisecond(ms)time slots which are then multiplexed together to form data frames. Each packet transmitted within a time slot contains 260 bits of information.

It is desirable that remote receivers reliably receive the transmitted information in each packet. However, due to a variety of factors, including environmental conditions, the messages contained in some packets are not always successfully received.

FIG. 1 is a graph showing the condition of an FM signal 12 at the receiver location. Signal 12 has spatial variations in signal strength (i.e., burst errors or nulls 14) that occur for discrete periods of time. Nulls 14 represent portions of signal 12 having a substantial loss of signal strength. Information in signal 12 coinciding with nulls 14 will not be successfully received by the target receiver.

A string of message packets 16, as described above in Gaskill, are shown extending along a horizontal axis representing time. Individual packet 18 of packet string 16 reaches the receiver during null 14. The null 14 destroys some or all of the bits in packet 18.

To correct for unsuccessfully received bits, the system in Gaskill includes a block error checking and correction code (ECC) scheme. However, the ECC scheme in Gaskill, can only correct for a limited number of corrupted bits in each packet (e.g., 7%).

To increase the probability of successfully receiving messages, the pager system in Gaskill retransmits the same message several times in each frame. However, retransmitting messages burdens a valuable communication resource, namely, the transmission path bandwidth. Each time a message is retransmitted, an additional portion of the transmission bandwidth is used for the same message instead of first transmission of other messages.

Another problem with simply retransmitting messages is that the burst error that corrupted the first message may also corrupt subsequent transmissions of the same message.

Several techniques have been devised for reducing the effects of nulls in transmission signals. For example, the same message can be transmitted over multiple frequencies. Since drop-out characteristics change according to carrier frequency, it is likely that portions of corrupted messages transmitted at a first frequency could be successfully received at an alternate carrier frequency.

Multiple transmitter stations are located at different physical locations so that the physical origin and signal strength of the message sent from each transmitter is different. The drop-out characteristics for the signals sent from each transmitter station are likely to be different. Thus, it is likely that the message will be successfully received from at least one of the multiple transmitter stations.

Transmitting and receiving the same message at different frequencies, or transmitting the same message from multiple transmitter stations, requires complex transmitter and receiver circuitry making the communication system more expensive to manufacture and operate.

Another technique for reducing the effects of burst errors involves interleaving multiple message packets together thus creating better burst error correction capabilities. Because receivers are portable, the signal drop-out characteristics at the receiver often change. As will be described below, transmitting a single interleaved packet size for varying signal drop-out conditions is not completely effective in minimizing burst error effects.

Accordingly, a need remains for increasing the probability of successfully receiving messages to receivers without using addition signal bandwidth.

SUMMARY OF THE INVENTION

Packets are transmitted in different block sizes according to the speed of motion of the receiver. The packet block size is selected to minimize the effects of burst errors (i.e., nulls) caused by multipath. Thus, the proportion of individual packets corrupted by nulls are reduced thus increasing the probability that each packet will be successfully received.

At relatively slow speeds, nulls in the transmitted signal are, in general, wide and have a relatively long time period before occurrence of the next null. At higher receiver speeds, nulls at the receiver are narrower and have a shorter time period. The size of packet blocks are adjusted according to these varying receiver null or drop-out characteristics so that only one null is likely to occur during the transmission of any one packet block. Thus, any corrupted data is distributed over an appropriate number of packets for the duration of the null.

Packets are encoded into packet blocks by first interleaving the packets together in a register. The interleaved packet block is then transmitted to the receiver. Because the packets are interleaved, a single burst error is dispersed over multiple packets. The receiver can then decode the packet block into the original packets. Since each packet now contains only a small proportion of the burst error, standard ECC schemes can be used to correct for any packet bit errors.

The transmitter determines the speed of the receiver, and accordingly the packet block size using various techniques. For example, an expected receiver speed is determined according to either the subject matter of the transmitted message, the type of receiver or via a two-way communication system where the receiver transmits receiver speed directly back to the transmitter.

Thus, the size of the transmitted packet blocks are varied to minimize the effects of burst errors, in turn, increasing the probability that packets will be successfully received without using any additional signal bandwidth or additional transmitter or receiver circuitry.

The foregoing and other objects, features and advantages of the invention will become more readily apparent from the following detailed description of a preferred embodiment of the invention which proceeds with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a graph showing signal drop-out characteristics for a receiver moving at a relatively slow speed.

FIG. 2A is a schematic diagram showing a communication system according to the invention, that changes the size of transmitted packet blocks according to the speed of motion of the receiver.

FIG. 2B is a detailed hardware schematic of the communication system shown in FIG. 2A.

FIG. 2C is a step diagram showing the various hardware operations performed by the communication system shown in FIG. 2B.

FIGS. 3-5 are graphs showing signal drop-out characteristics for receivers moving at 15 miles per hour, 30 miles per hour and 60 miles per hour, respectively.

FIG. 6 is an enlarged schematic diagram showing a portion of a noninterleaved packet string shown in FIGS. 3-5

FIG. 7 is an enlarged diagram of a burst error previously shown in FIG. 3.

FIG. 8 is a schematic diagram showing portions of the noninterleaved packet string corrupted by the burst error in FIG. 7.

FIG. 9 is a schematic diagram of a transmitter register used for encoding packets into a variable sized packet block according to the invention.

FIG. 10 is a schematic diagram showing a portion of the encoded packet block in FIG. 9 corrupted by the burst error in FIG. 7.

FIG. 11 is a schematic diagram of a receiver register for deinterleaving encoded packet blocks.

FIG. 12 is a graph showing the drop-out characteristics for a receiver moving at a fast travel speed.

FIG. 13 is a schematic diagram of the transmitter register shown in FIG. 9 encoding a new packet block size according to the drop-out characteristics shown in FIG. 12.

FIG. 14 is a schematic diagram showing burst errors in the small packet block encoded in FIG. 13.

FIG. 15 is a schematic diagram of the receiver register after receiving the packet block in FIG. 14.

FIG. 16 is a schematic diagram of a two-way communication system according to a second embodiment of the invention having a receiver that transmits travel speed of the receiver back to the transmitter.

FIG. 17 is a detailed hardware schematic for the system shown in FIG. 16.

FIG. 18 is a step diagram showing the operations performed by the communication system in FIG. 16.

DETAILED DESCRIPTION

FIG. 2A is a schematic diagram showing a transmitter 20 that sends variable sized packet blocks according to the speed of motion of the receiver. A human 22 wears a paging receiver 24 at a wrist location and represents an object that is stationary or moving at a relatively slow speed. For example, the typical walking speed of a human is approximately three miles per hour (MPH). A car 28 carries a receiver 29 and represents a object that travels at a relatively high speed. For example, car 28 typically travels at between 30 and 60 MPH or at walking speeds in stop and go traffic.

At low receiver speeds (e.g., 3 MPH) the transmitter sends a relatively large packet block 21 and at higher receiver speeds (30 MPH-60 MPH) the transmitter sends a smaller packet block 26. A large packet block size is defined as being encoded using a relatively large number of message packets. A small packet block size is defined as being encoded using a relatively small number of packets. Varying the packet block size according to the speed of motion of the receiver increases the probability that each packet in the packet block will be successfully received during burst error conditions.

FIG. 2B is a detailed schematic of the transmitter 20 shown in FIG. 2A. The transmitter includes a receiver/register 92 for receiving and temporarily storing a message for transmission to a receiver. The message is transferred to the transmitter either over a conventional land line or via wireless transmission. A central processing unit (CPU) 94 determines the expected receiver travel speed according to the message content or the type of receiver as will be discussed in detail below.

The message is then transferred to a register 96/98. The register 96 interleaves a message into a large packet block size. The register 98 interleaves a message into a small packet block size. The receiver/register 92 feeds the message either to register 96 or register 98 according to the expected receiver travel speed determined by CPU 94. A transmitter 100 then sends either the large packet block or the small packet block to the receiver.

FIG. 2C is a step diagram showing the operations performed by the receiver shown in FIG. 2B. The transmitter receives a message for transmission in step 72. The transmitter then determines the expected receiver travel speed in step 74.

The system in FIG. 2A does not provide two-way communication. Therefore, the transmitter determines the expected travel speed of the receiver by either the type of (e.g., wristwatch pager, car receiver, etc.) receiver or the contents of the transmitted message. For example, if the message is being transmitted to a wrist pager, the expected travel speed is slow. However, if the message is being transmitted to a car, the expected receiver travel speed is faster.

Alternatively, the contents of the message can determined according to the contents of the transmitted message. For example, traffic information is likely to be received by a receiver located in a car. Therefore, the expected receiver travel speed will be relatively fast.

If the expected receiver travel speed is fast, decision step 76 jumps to step 78 where the transmitter assembles packets into a relatively small packet block size. If the expected receiver travel speed is slow, decision step 76 jumps to step 80 where the transmitter assembles the packets into a relatively large packet block size. The packet blocks are then sent from the transmitter to the receiver in step 82.

To explain further, FIGS. 3-5 are graphs showing signal drop-out characteristics for receivers moving at 15 MPH, 30 MPH and 60 MPH, respectively. The graph in FIG. 3 shows the same signal drop-out characteristic previously shown in FIG. 1. The vertical axis of FIGS. 3-5 represent signal strength in decibels (db) and the horizontal axis represents time. The individual packets in packet string 16 are encoded and transmitted in a manner similar to that discussed in Gaskill above which is herein incorporated by reference.

Null 14 represents a burst error that typically occurs from destructive interference due to signal reflections. Signal drop-out characteristics vary according to signal strength, receiver sensitivity and other environmental conditions. The signal strength, receiver characteristics and other environmental conditions associated with FIGS. 3-5 are assumed to be substantially similar. The physical condition that has varied the drop-out conditions between FIGS. 3, 4, and 5 is the speed of motion of the receiver. For example, the speed of motion of the receiver experiencing the drop-out conditions in FIG. 3 is 15 MPH and the speed of motion of the receiver experiencing the drop-out condition in FIG. 4 is 30 MPH.

The vertical location of packet string 16 defines a threshold signal strength level. When the signal strength of signal 12 is above packet string 16 the message signal 12 is likely to be successfully received by the receiver. When the signal strength of signal 12 falls below packet string 16 the message on signal 12 is destroyed by null 14 and not successfully received by the receiver.

When the receiver is moving at 15 MPH (FIG. 3), nulls 14 have a relatively wide time duration 36 and have a relatively long time period 30 between adjacent nulls. Alternatively, when the speed of motion of the receiver is at 60 MPH (FIG. 5), nulls 34 are narrow and have a relatively short time period 40.

Referri