An asynchronous sampling monitor for optical DPSK signals by using a two-tap histogram is proposed. By means of statistical analysis, a parameter is defined to measure the Q-factor to indicate the OSNR degradation. Compared to traditional monitoring methods, this method requires no synchronisation before sampling and is thus cost-effective.

A new fault-branch detection scheme is proposed to troubleshoot the breaks of any distribution fibers in a timedivision multiplexing (TDM) passive optical network. We employ a continuous optical frequency sweeper at the optical line terminal (OLT) and an interferometric (IF) device at each optical network unit (ONU). By analyzing the spectrum of the returned combined signals at the OLT, we can obtain the status of all branches. This detection method not only uses a small optical frequency band for surveillance monitoring, but is also simple to operate. Furthermore, a modified architecture is proposed to relax the specifications of IF devices. The tolerance of the IF device length was analyzed using the Monte-Carlo simulation method.

reduced. As shown in Figure 7, the peak directivity of this case is less than 4 dBi at 5.80 GHz (compared with more than 7 dBi in the Fins PNRE case). CONCLUSIONAn impedance-matched, 5.8-GHz extruded-fin heatsink antenna was introduced having two orthogonal fin orientations. The orientation was shown to have a significant effect on the antenna directivity and gain performance across a frequency range. Thus, the extruded-fin heatsink provides the flexibility of either a highor low-directivity antenna simply by the orientation of the fins. The best choice will be determined by the specific application.Furthermore, the results demonstrated that the extruded-fin heatsink can improve the antenna performance while reducing the component count and providing a mechanism for removing thermal energy from high-power electronics such as an RF power amplifier. These performance improvements allow the use of lowcost, lossy FR4 substrates (which would typically be considered a poor substrate choice) with high dielectric losses. FL, March 19 -22, 2007, pp. 419 -422. 3. C.A. Balanis, Antenna theory: Analysis and design, 3rd ed., Wiley, Hoboken, NJ, 2005, pp. 641-642. 4. T.A. Milligan, Modern antenna design, McGraw-Hill, New York, NY, 1985 Because BER is the most preferred parameter for fault management, its monitoring techniques are gaining importance. One such signal quality monitoring scheme is based on the Q-factor monitoring, which can be obtained from an amplitude histogram obtained using synchronous sampling [6,7]. However, this method requires clock recovery to ensure synchronization in the sampling process. In practice, clock recovery electronics are both bit-rate and signal-format dependent. To reduce system complexity, Shake et al. used the amplitude histogram generated by asynchronous sampling to calculate the averaged Q-factor (Q avg ) without clock recovery [8,9]. With this scheme, bit-rate and signal-format transparency can be achieved. Furthermore, the Q avg value was found to have a highly linear relationship with the real value of Q-factor (Q t ) [10]. Photonic sampling methods can be classified into two types: all-optical and electro-optical (EO) sampling. The former gates the signal with optical sampling pulses, and the gating process is usually based on second-order or third-order nonlinear optical effects [11,12]. The latter utilizes electrical pulses to drive an optical sampling gate, which is often an EO or electroabsorption (EA) modulator [13]. Because the modulators and electrical pulse generators are relatively simple and low in cost, EO sampling is widely used in high-speed sampling modules. The amplitude histogram depends on the sampling uniformity, which is related to the relationship between the signal rate (f s ) and sampling rate (f c ). This histogram is also affected by the sampling position. For asynchronous sampling with a small number of sampling points, the histogram cannot represent the real statistics of the original signal if f s is very close to an integer multiple of f c . In th...

No abstract

No abstract