High Speed Error Detector for Fading Propagation Channels
Developing a network protocol over a propagation channel requires a detailed knowledge of the characteristics of the channel. As a function of time, the channel may be impaired by noise if the channel experiences power loss due to fades that result from a variety of physical effects. The bit error ratio is the system-level characterization that reveals the effectiveness of the channel for communications.
The problem with available bit error rate testers is that their results will be dominated by fades if the fades occur more rapidly than the instrument can be polled for data. With propagation channels such as those provided by free space optical communications, the data rate is many gigabits per second, and in the narrow field of test equipment available, the fades of the channel are a dominant feature of the link. Frequent fades can significantly influence the measurement results.
The invention is a hardware design that overcomes the limitations of available bit error rate testers. The design comprises the following unique elements:
• The receiver has a programmable parameter that directly determines the time for synchronization by identifying the amount of pattern to be matched before synchronization is declared. Commercial error detectors offer no detail on this process.
• The receiver has a programmable parameter that determines the amount of pattern mismatch before it declares loss of synchronization, and the exact timing of synchronization loss is provided to the bit level.
• An onboard clock drives a process to determine the number of bits lost during a total loss of the channel. Commercial error detectors do not provide this measurement.
• Large counter field widths eliminate counter overflows, and binary mode result polling is rapid so that time-dynamic results are not lost due to channel disturbances.
• A bank of counters totals the number of consecutive bits that have been received error free. These counters provide an indication of the time statistics of the channel when data and packets can be piped through before fades occur. Such data can immediately reveal the maximum optimal packet length that can be sent over the channel.
• A bank of counters records the frequency of fade duration.
• Separate counters are included to measure total bits sent, total bits lost due to loss of synchronization, and bits errored while synchronized. These counters allow calculation of effective bit error rates, thus eliminating the domination of the bit error rate statistics by sync-out conditions. These counters can be used to calculate the link availability and the estimated best possible instantaneous data throughput of the link.
• High-speed digital pulse outputs indicate synchronization and bit error status. These signals can be digitized to provide further insight into the link behavior.
• The statistics measured by the invention provide information to calculate the frequency of fades and the average time between fades.
Thus, the invention, for which a working prototype has been developed, allows for more precise characterization of the bit error rate of a communications link that has extensive fade and drop-out conditions, such as those that are found in free space optical communications.
Patent Status: U.S. patents pending.CONTACT:
Mr. M. T. Hickman