Recently, Nutaq performed an interesting demo by plugging a Nutaq Perseus board with an ADAC250 FPGA mezzanine card (FMC) into a wideband RF receiver. The wideband receiver has a maximum instantaneous bandwidth of 100 MHz and is able to cover frequencies from 100 kHz to 20 GHz. Plugging in the Perseus board with ADAC250 FMC completes the loop by adding the digital signal processing part of the system. This makes it the perfect tool for spectrum monitoring and signal intelligence (SIGINT) applications. In this blog, we will introduce one of those SIGINT technologies, one that is already a major part of our lives but not well known: spread-spectrum.
Spread-spectrum systems use a spreading algorithm based on a pseudo-noise (PN) code to spread a radio signal across a wide bandwidth. The result is a signal that is essentially buried in the noise floor of the radio band. On the receiver side, this wide bandwidth needs to be correlated to despread the signal. The correlation process also causes any other received signals to be spread as the wanted signal is despread. This causes any unwanted signals to be seen as noise. The end result is a signal that is difficult to detect, does not interfere with other services, and still carries a great bandwidth of data. It also becomes hard to intercept, to jam, and to demodulate, which is why this technology was used in military applications for so many years.
There are two main ways to implement spread-spectrum technology: direct sequence and frequency hopping.
The generation of direct-sequence spread-spectrum signal (spreading) is shown in Figure 1:
Figure 1: Direct-sequence spread-spectrum signal (spreading operation)
The narrow-band signal and the spread-spectrum signal use the same amount of transmit power and carry the same information around the same carrier frequency. However the power density of the second signal (spread spectrum) is much lower. This capability is the main reason for all the interest in spread spectrum. At the receiving end, the spread-spectrum signal is despread to generate the original narrow-band signal, as shown in Figure 2.
Figure 2: Direct-sequence spread-spectrum signal (despreading operation with interferer)
In frequency hopping, as its name implies, the carrier frequency ‘hops’ or changes in time, as shown in Figure 3:
Figure 3: Frequency hopping generated signal
Frequency hopping does not spread the signal as compared to the direct sequence method, so there is no processing gain. The processing gain is the increase in power density when the signal is despread and it improves the received signal’s signal-to-noise ratio (SNR). In other words, more power is needed with frequency hopping to achieve the same performance compared to the direct-sequence method. Also, at the receiver, this makes synchronization more difficult because both frequency and time synchronization is needed. This tends to increase the hardware requirements for the radios (which means more money). Also, the carrier frequency hops almost guarantee interference with other radios due to the narrow-band signal, thus resulting in a higher bit error rate (BER).
As a result, frequency hopping is more popular for voice applications. Frequency hopping does, however, react better in multipath fading environments and can usually carry more data owing to its narrow-band signal. Regarding interference immunity, the action of hopping will make the signal avoid specific interference from time to time. Frequency hopping is more popular than direct-sequence – it’s actually the only way to survive in the 2.45 GHz band because of the leakage from microwave ovens which can exceed 10W.
Whether you use direct-sequence or frequency hopping, spread-spectrum technologies offer interesting advantages. Many solutions, including the Nutaq wideband receiver, are available to engineers when developing with applications this technology. Some applications already implement spread spectrum, including:
• Wireless local area networks (WLAN)
• Global Positioning Systems (GPS)
• Space systems
• Long-range communications