Agilent Measurement Journal - Issue 5 - 2008 - (Page 41)

Strategies for searching There are many possible algorithms to search for the interval [x–, x+]. There are several items to consider during the implementation of a fast TJ measurement using the bracketing approach: • Because measurement times increase with decreasing Pe, the search should be performed from left to right for the left edge, and from right to left for the right edge. • The main goal of the search algorithm is to minimize the number of failed attempts to ﬁnd x+. At 10 Gb/s it takes about ﬁve minutes to compare 3e12 bits, the longest measurement time during each iteration. • Once x– has been determined, x+ is often within ∆x because of resolution limitations. • The search can be optimized if it is realized as an iterative process. The resolution can be ﬁne tuned continuously from a coarse step to a ﬁne step. • The measurement times can be optimized for a given measurement accuracy. • To get better initial values for the search, it is a good idea to perform a relatively fast complete bathtub scan. From x+, either directly or by ﬁtting an inverse error function to the data. • If the device under test has a Pe ﬂoor, the search may be stuck since the Pe never gets below 10-12. This can be accounted for with a robust implementation. are given in Tables 1 and 2. Note that measurements times with the bracketing approach are not strictly repeatable, since at lower Pe values the time until the ﬁrst error is observed is randomly distributed. Table 1. Measurement times for 10 Gb/s signals; time resolution set to 1 ps RJ = 0 ps RJ = 2 ps RJ = 4 ps RJ = 6 ps DJ = 0 – 16 min 17 min 17 min DJ = 15 ps 10 min 12 min 16 min 26 min DJ = 30 ps 10 min 10 min 12 min – Table 2. Measurement times for 10 Gb/s signals; time resolution set to 5 ps RJ = 0 ps RJ = 2 ps RJ = 4 ps RJ = 6 ps DJ = 0 – 10 min 10 min 11 min DJ = 15 ps 10 min 10 min 14 min 8 min DJ = 30 ps 10 min 10 min 10 min – using a linear search algorithm for various combinations of RJ and DDJ.8 Average measurement times for this implementation The data in the tables above are obtained through simulations, which were carried out using the average time between errors.8 When the RJ values were low, the measurement was completed in 10 minutes, independent of DJ. This is because the slope of the bathtub curve is directly related to the value of RJ. In the low Pe region, it takes only one data point to bracket the 10-12 point. The measurement time is then dominated by the time required to compare 2.996e12 bits (ﬁve minutes), once per slope. This is independent of the delay resolution used. The minimum test time at 10 Gb/s is always 10 minutes, no matter how coarse the resolution. Agilent Measurement Journal 41

Table of Contents Feed for the Digital Edition of Agilent Measurement Journal - Issue 5 - 2008

Agilent Measurement Journal Issue 5 Finding a New Path when Assumptions Break Down Contents Testing MIPI D-PHY Protocols Oscilloscope Firmware Update DNA Replication Joining the LTE/SAE Trial Initiative Testing Proteins Praise for DC Power Analyzer Scripting Features for AMDS Testing Hybrid PCBA Systems Improving the Gene Expression System Workflow Turning a “Good Enough” Test Strategy into One That’s Reliable, Repeatable and Traceable Understanding the Effects of Limited-Bandwidth Channels on Digital Data Signals Introducing the 3GPP LTE Downlink Validating the 26 Validating the Physical and Protocol Layers in DDR Memory Interfaces Understanding Total Jitter Measurements of Low Probabilities Using Behavioral-Model Simulation to Accurately Predict First-Order PLL Performance Creating Synchronous High-Frequency Sampling across Multiple Digitizers Overcoming the Challenges of Testing FlexRay Networks Understanding Operating Life and Repeatability of Electromechanical Switches and their Effect on Total Cost of Ownership Implementing Micro and Nano LC/MS Techniques for High Sensitivity Lipidomics Analysis

Agilent Measurement Journal - Issue 5 - 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.