Additional I/O Connections; Ethernet Command And Control; Noise Blankers; Noise Reduction - Apache Labs ANAN-7000DLE MKII User Manual

1

Additional I/O Connections:

RCA - Line In, PTT In, PTT Out, 2 – DIG inputs
DB9F - 7 Software Configurable Open Collector Outputs
SMA – XVTR output, 10Mhz Reference Input
3.25mm Barrel Mic, CW Key, Headphones and two 6.25mm Speaker Outputs
RJ45 - Ethernet LAN Connector(SDR)
RJ45 Ethernet LAN Connector (Embedded PC version)
HDMI (Embedded PC version)
2 x USB 3.0 Gen2 (Embedded PC version)
USB-C (Embedded PC version)
Wi-Fi SMA Antenna Connector (Embedded PC version)

Ethernet Command and Control:

The ANAN Transceivers use a Gigabit Ethernet interface to connect to your LAN. Ethernet command and control
features huge bandwidth, better noise isolation from the PC, and networked radios with remote access. Beta
testers report good performance across 802.11n WiFi links.

Noise Blankers:

The PowerSDR provides a choice of two Wideband Noise Blankers in addition to the new Spectral Noise Blanker
by Warren NR0V.
The two Wideband Noise Blankers are:
A pre-emptive Blanker which effectively slews it's output to zero before an impulse arrives, and then
slews back to full amplitude after the impulse passes.
An Interpolating Noise Blanker which is also pre-emptive, but has modes such as Linear interpolation of
the signal during the impulse. Wideband Noise Blankers, while they are the best choice in some
situations, have a well-known issue that they may become ineffective in the presence of strong adjacent
signals, for example during a busy contest. The Spectral Noise Blanker cleverly overcomes this
deficiency by using Linear Predictive Coding (LPC). LPC provides the capability to predict a sequence of
samples by analyzing the spectral content of the samples before and after the sequence. By comparing
the predictions with the measured samples, impulses are detected and blanked or reduced. The LPC is
again used to predict what the signal waveform should be during the incident pulse and thereby replace
the corrupt samples with clean signals.

Noise Reduction:

Two types of Noise Reduction algorithms are provided to minimize random noise.
The first is a special implementation of a Least Mean Square (LMS) algorithm and the second is a new Spectral
Noise Reduction NR2 (2015). LMS algorithms are used in most Software Defined Radio's and Digital Signal
Processing modules due to their relatively simple implementation and low compute requirements. However they
use only the input signal as a reference to identify the output and therefore are unable to achieve optimal
signal-to-noise ratios and they sometimes yield an unusual "in the barrel" or "underwater" sound. The new
Spectral Noise Blanker specifically uses sophisticated statistical models of speech and noise to produce superior
signal-to-noise ratios and vibrant sound output.
Copyright Apache Labs © page 8 26 February 2019