Error Rates of Digital Signals in Charge Transfer Devices

As a packet of charge is transferred through a charge transfer device (CTD), the size of the packet is altered owing to effects of incomplete transfer1-2 and noise. 3-9 At the output the size of each packet is measured and, depending on its size, a decision is made as to the initial size of the packet. Usually the decision will be correct. However, occasionally the cumulative effects of incomplete transfer and noise will result in a sufficiently distorted charge packet that an error will be made. It is the purpose of this paper to calculate the probability of making such a detection error. When this probability is multiplied by the rate of detection (the clock frequency), we obtain 1795 1796 T H E B E L L SYSTEM T E C H N I C A L J O U R N A L , D E C E M B E R 1973 the error rate for a single device. Multiplying by the number of devices of interest in the storage unit or in the processing unit, we obtain the total error rate, a very useful quantity for evaluating digital systems. (By "detection," we include regeneration; by a "single device," we mean a single unregenerated line of transfer elements.) To calculate the probability of detection error we assume that the effect of incomplete transfer on the signal can be treated in terms of the usual small signal analysis.10,11 (The coefficient of incomplete charge transfer, a, is assumed to be constant, independent of the size of the signal.) Charge gain or loss because of leakage current is assumed to be sufficiently small that it can be ignored.